Search

Fresnel incoherent correlation holography (FINCH) can achieve high-precision and non-scanning 3D imaging. However, as a holographic imaging technology, the huge bandwidth requirements and the amount of holographic data transmitted have always been one of the important factors limiting its application. In addition, the hardware cost of pixel array-based CCD or CMOS imaging is very high under high resolution or specific wavelength conditions. Accordingly, a single-pixel Fresnel incoherent correlation holography (SP-FINCH) compressed imaging method is proposed, which replaces pixel array detector with single-pixel detector and designs a Trumpet network to achieve low-cost and high-resolution imaging. Firstly, a modified FINCH imaging system is constructed and data acquisition is carried out using a single-pixel detector. Secondly, a Trumpet network is constructed to directly map the relationship between one-dimensional sampled data and two-dimensional image in an end-to-end manner. Moreover, by comparing the reconstructed images using neural network with that using commonly used single-pixel reconstruction methods, the results indicate that the proposed SP-FINCH compressed imaging method can significantly improve the quality of image reconstruction at lower sampling rate and achieve imaging without phase-shifting operation. The proposed method has been shown to be feasible and advantageous through numerical simulations and optical experiment results. [ABSTRACT FROM AUTHOR]

Non-blind image deblurring has attracted a lot of attention in the field of low-level vision. However, the existing non-blind deblurring methods cannot effectively deal with a saturated blurry image. The key point is that the degradation model of saturated blurry images does not satisfy the linear convolution model of a conventional blurry image. To solve the problem, in this paper, we proposed a novel deep non-blind deblurring method, dubbed saturated image non-blind deblurring network(SDBNet). The SDBNet contains two trainable sub-network, i.e., confident estimate network (CEN) and detail enhance network (DEN). Specifically, the SDBNet uses CEN to estimate the confidence map for the saturated blurry image, which is used to recognize saturated pixels in the blurry image, and then uses the confidence map, and blur kernel to restore the blurry image. Finally, we use DEN to enhance the edges and textures of the restored image. We first pre-train CEN and DEN. In order to effectively pre-train CEN, we propose a new robust function, which is used to generate label data for CEN. The experimental results show that compared with several existing non-blind deblurring methods, SDBNet can effectively restore saturated blurry images and better restore the texture, edge, and other structural information of blurry images. [ABSTRACT FROM AUTHOR]

The key challenge of image layer separation stems from recognizing different components in a single image. Typical methods optimize the modeling of different components by performing low-level supervision on the separated image to minimize its per-pixel difference from the groundtruth, which relies on substantial training samples to learn diverse components robustly and avoid overfitting spurious coupled patterns. In this work, we perform supervision on the contrastive distribution between the predicted separated images. Specifically, our proposed method separates components in parallel and seeks to maximize the distribution consistency between the separated components' contrast and their corresponding groundtruth contrast in the latent space. Such supervision pushes the model to focus on contrastive modeling between different components of the input image. Besides, the learned latent representations of different components directly guide the weight optimization of convolution kernels in the decoder, which achieves more comprehensive separation than traditional skip connection while reconstructing target images by the decoder. We validate the effectiveness and generalization of our CFDNet on two single image layer separation tasks, including image reflection separation and intrinsic image decomposition. Extensive experiments demonstrate that our CFDNet consistently outperforms other state-of-the-art methods specifically designed for either of image separation applications. [ABSTRACT FROM AUTHOR]

Weakly Supervised Semantic Segmentation is a crucial task in computer vision. However, existing methods that utilize Class Activation Maps (CAMs) with classification tasks can only identify a small part of the region. To address this limitation, we propose a novel Attention Activation Remodulation (AAR) scheme that leverages traditional CAMs and the remodulation branch to obtain weighted CAMs for recalibrated supervision. The AAR scheme re-arranges important features' distribution from the channel and space perspectives, which regulates segmentation-oriented activation responses. In addition, we propose a Feature Pixel Extraction Module (FPEM) that utilizes contextual information to improve pixel prediction. Furthermore, the proposed scheme can be combined with other methods to improve overall performance. Extensive experiments on the PASCAL VOC 2012 dataset demonstrate the effectiveness of the AAR mechanism and FPEM module. [ABSTRACT FROM AUTHOR]

Masked Autoencoders (MAE) have been popular paradigms for large-scale vision representation pre-training. However, MAE solely reconstructs the low-level RGB signals after the decoder and lacks supervision upon high-level semantics for the encoder, thus suffering from sub-optimal learned representations and long pre-training epochs. To alleviate this, previous methods simply replace the pixel reconstruction targets of 75% masked tokens by encoded features from pre-trained image-image (DINO) or image-language (CLIP) contrastive learning. Different from those efforts, we propose to Mimic before Reconstruct for Masked Autoencoders, named as MR-MAE, which jointly learns high-level and low-level representations without interference during pre-training. For high-level semantics, MR-MAE employs a mimic loss over 25% visible tokens from the encoder to capture the pre-trained patterns encoded in CLIP and DINO. For low-level structures, we inherit the reconstruction loss in MAE to predict RGB pixel values for 75% masked tokens after the decoder. As MR-MAE applies high-level and low-level targets respectively at different partitions, the learning conflicts between them can be naturally overcome and contribute to superior visual representations for various downstream tasks. On ImageNet-1K, the MR-MAE base pre-trained for only 400 epochs achieves 85.8% top-1 accuracy after fine-tuning, surpassing the 1600-epoch MAE base by + 2.2 % and the previous state-of-the-art BEiT V2 base by + 0.3 %. Pretrained checkpoints are released at https://github.com/Alpha-VL/ConvMAE. [ABSTRACT FROM AUTHOR]

We propose a novel network to perform auxiliary-free video matting task. Unlike most existing approaches that require trimaps or pre-captured backgrounds as auxiliary inputs, our method uses binary segmentation masks as priors and realizes the auxiliary-free matting. Furthermore, we design the attention-based memory block by combining the idea of the memory network and self-attention to compute pixel-level temporal coherence among video frames to enhance the overall performance. Moreover, we also provide direct supervision for the temporal-guided memory module to boost the network's robustness. The validation results on various testing datasets show that our method outperforms several state-of-the-art auxiliary-free matting methods in terms of the alpha and foreground prediction quality and temporal consistency. [ABSTRACT FROM AUTHOR]

In reversible data-hiding techniques, the quality of the steganographic image and its embedding capacity are the most crucial characteristics. The main objective of this study is to enhance the Biswapati et al. approach, which embeds secret data directly in interpolated pixels without taking context pixel properties into account. Additionally, they recorded an extremely large position value of the weighted matrix, which led to a significant amount of visual distortion. To deal with this issue, a novel reversible data-hiding method based on a multilayer center folding technique (MCFT) is developed. The suggested approach divides the interpolated cover image into 5 × 5 non-overlapping blocks and then sorts them in descending order of standard deviations. As a result of this sorting, edges and textures are more effectively preserved while also reducing the appearance of frequent interpolation defects. Unlike previous weighted matrix approaches, the position value of the weighted matrix is not embedded directly to generate the stego-image. MCFT is used before embedding to reduce the difference between an image pixel and a stego pixel so that image quality is not destroyed. The Huffman encoding technique is used to preprocess the acquired secret data to increase the embedding capacity as much as feasible. Additionally, to increase embedding rates, dual images are used for exploiting the characteristics of exploiting modification direction. The findings achieved significantly outperform state-of-the-art algorithms, which provide visual quality of more than 50.8 dB peak signal-to-noise ratio PSNR. When compared to modern techniques, the approach also offers the highest level of security and can retrieve the original image without losing any data. [ABSTRACT FROM AUTHOR]

Recent research has shown the vulnerability of deep networks to adversarial perturbations. Adversarial training and its variants have been shown to be effective defense algorithms against adversarial attacks, enhancing the defense abilities of deep neural networks by training them to fit adversarial examples. However, the significant computational burden of generating strong adversarial examples has rendered the process time-consuming, presenting a challenge for efficient training. In this paper, we propose adversarial training with robust area (ATRA), a highly efficient variant of adversarial training. We experimentally find that certain pixels in the image play a crucial role in improving robust accuracy, which we refer to the collection of discrete pixels as the high-robust area. Based on the robust area of the input instance, ATRA generates adversarial examples by applying an adaptive perturbation. Furthermore, we investigate the transferability of the high-robust area during the attack iteration process and experimentally demonstrate its effectiveness. Therefore, ATRA has the advantage of reducing the additional cost of generating strong adversarial examples while maintaining model robustness. Our experimental results on MNIST, CIFAR10, and TinyImageNet show that our method outperforms current state-of-the-art baselines with significantly less additional training time required, especially on MNIST where our method requires 18 × less training time. Furthermore, our method also achieves good performance under different adversarial attacks such as FGSM, CW, and AutoAttack. [ABSTRACT FROM AUTHOR]

Pixel value ordering (PVO) and prediction error expansion (PEE) techniques are now being combined for improved performance in the domain of reversible data hiding (RDH). Conventional PVO can only embed two data bits in a smooth picture block in a row or column, which is insufficient to meet the contemporary demands of much payload in RDH. There is a lot of room to enhance the bit stuffing by inserting more than two data bits in a smooth block while maintaining acceptable visual image quality and data security. In this study, we suggest a context pixel-based data concealing strategy based on PVO combined with PEE, in which each block can insert two or more data bits. This proposed approach has three distinct features: smooth block selection, minimum and maximum position selection, and pixel difference with median value for data embedding. Furthermore, more than two pixels can be used to embed the secret data bit. The block is recommended by the selection of the neighbouring and present block's pixel. The use of context pixels in context pixel-based PVO (CPPVO) improves both embedding efficiency and picture quality. When the block size is Z = (3 × 3) , the average embedding capacity (EC) of our novel approach is over 70,000 bits, and the average PSNR is over 40 dB. The results of the experiments reveal that our technique is more dependable than other state-of-the-art systems in terms of security, picture quality, and secret volume. [ABSTRACT FROM AUTHOR]

The feature matching algorithm based on deep learning has achieved superior performance compared to traditional algorithms in terms of both matching quantity and accuracy, but there are still some high-error matching results in complex scenes, which adversely affects the subsequent work. Based on SuperGlue, we propose an accurate feature matching algorithm via outlier filtering. Firstly, DBSCAN real-time superpixel segmentation (RTSS-DBSCAN) is used to divide the image into regions, and then the outlier filtering module is designed according to the local similarity principle of feature matching. On the premise of not affecting the correct matching results, the matching results with high errors will be filtered to improve the matching accuracy. Meanwhile, due to the lag of traditional Exponential Moving Average algorithm (EMA), an adaptive EMA is designed and integrated into the SuperGlue training process to further improve the training speed and matching accuracy. We evaluate the overall performance of the matching method using the AUC of pose error at the thresholds (5°, 10°, 20°), a common evaluation metric, to provide a more detailed and intuitive evaluation of the matching effectiveness using precision and recall. The experimental results show that the method in this paper can effectively filter the matching results with large errors and has high accuracy and robustness. The AUC of pose error at thresholds (5°, 10°, 20°) reaches 36.53, 56.23, and 73.68, and the precision and recall reach 80.07 and 91.52, respectively, which have better matching results compared with other algorithms. [ABSTRACT FROM AUTHOR]

We study the generation of adversarial examples to test, assess, and improve deep learning-based image quality assessment (IQA) algorithms. This is important since social media platforms and other providers rely on IQA models to monitor the content they ingest, and to control the quality of pictures that are shared. Unfortunately, IQA models based on deep learning are vulnerable to adversarial attacks. Combining the characteristics of IQA, we analyze several methods of generating adversarial examples in the classification field, and generate adversarial image quality assessment examples by obtaining model gradient information, image pixel information and reconstruction loss function. And we create an adversarial examples image generation tool that generates aggressive adversarial examples having good attack success rates. We hope that it can be used to help IQA researchers assess and improve the robustness of IQA. [ABSTRACT FROM AUTHOR]

The aim of the paper is to introduce appropriate tools for quantifying the changes occurred at edge level when analyzing land cover changes. The research is conducted in two directions. Firstly, the binary change index at edge level is introduced and its relationship to the already existing binary change edge (computed at pixel level) is studied. It is proved that a certain inequality between the values of the two indices holds. The theoretical part is completed with real-world examples. The second direction investigated in the paper is the possibility of applying Markov type models, a basic tool in land cover change analysis, in quantifying the changes occurred at edge level. This combined approach is illustrated by a case study and it is shown that it provides complementary information related to the spatial extension of the land cover types and to their distribution. Thus, this type of analysis can improve the methodology of measuring the extent and the implications of land cover changes. [ABSTRACT FROM AUTHOR]

Digital images are information carriers in the latest era of technology. Therefore, securing digital images or keeping data, videos, or confidential documents during communication through the internet is challenging because an adversary always endeavors to acquire any related information about digital images or reserved data. Chaotic systems are immaculate for developing new cryptosystems due to their exceptional characteristics. In this paper, our primary focus is to introduce a procedure to generate unique secret key streams using a two-dimensional logistic map and create a sequence of positive integers with the Arnold Cat map. Confusion and diffusion are the two fundamental principles of any image encryption scheme. In our proposed image encryption scheme, confusion is done using a series of positive integers to devastate the correlation among the adjacent pixels of plain images in all directions, and diffusion is done with secret key streams. The consequences of the security and statistical analysis depict that the proposed image encryption scheme is safe against numerous attacks. [ABSTRACT FROM AUTHOR]

Image super-resolution (SR) is a classic problem of image processing. This paper proposes a self-learning interpolation method (SLIM) based on a single image by combining grid feature mapping with binary decision tree, which is not only transparent as the interpolation-based methods, but also achieves comparable performance as the learning-based methods. Firstly, it downsamples the given image I LR to obtain its low–low-resolution image I LLR , which is used to obtain sample data for the self-learning interpolation algorithm for enlarging I LLR to get I LR . Secondly, it provides a structural feature classification method to divide all of the samples into several groups, such that each class of I LLR is mapped to a matrix of coefficients for calculating the values of the pixels of I LR . The image I LR is approximated by executing the decision tree to refine the corresponding mapping matrix. Finally, the resulting high-resolution image I HR is obtained from the given image I LR by using the mapping matrixes. Experimental results show that SLIM achieves more smooth edges and better details on subjective vision than prevailing SR methods, and it is a transparent one but achieves comparable performances on PSNR and SSIM with the learning-based methods, while it outperforms the interpolation-based methods. It means that SLIM is both transparent and efficient and has much better subjective vision than other SR methods. [ABSTRACT FROM AUTHOR]

With the speedy progress of deep learning, text detection has received progressively increasing attention and considerable progress. The current mainstream approaches are usually based on instance segmentation to obtain the label of whether the pixel is text, as this can cope with arbitrary-shaped text. However, pixel-based prediction usually leads to overlapping neighboring texts, resulting in misdetection. To mitigate the above problems, we propose an approach to calculate text kernels and determine the attribution of boundary pixels. This way, all texts are labeled uniformly, facilitating model learning and effectively separating adherent texts. In addition, to cope with the complex and variable background of the text, we propose a practical feature enhancement module to handle it. The proposed module can explore different levels of features to represent text information of diverse sizes. Compared with current advanced algorithms, our method is competitive, which achieves the F1-measure of 87.3, 88.0, 82.8, 85.7, and 90.0 % on the ICDAR2015, MSRA-TD500, CTW1500, Total-Text, and ICDAR2013 datasets, respectively. [ABSTRACT FROM AUTHOR]

In this paper, we extend Open-set Semantic Segmentation (OSS) into a new image segmentation task called Generalized Open-set Semantic Segmentation (GOSS). Previously, with well-known OSS, the intelligent agents only detect unknown regions without further processing, limiting their perception capacity of the environment. It stands to reason that further analysis of the detected unknown pixels would be beneficial for agents' decision-making. Therefore, we propose GOSS, which holistically unifies the abilities of two well-defined segmentation tasks, i.e. OSS and generic segmentation. Specifically, GOSS classifies pixels as belonging to known classes, and clusters (or groups) of pixels of unknown class are labelled as such. We propose a metric that balances the pixel classification and clustering aspects to evaluate this newly expanded task. Moreover, we build benchmark tests on existing datasets and propose neural architectures as baselines. Our experiments on multiple benchmarks demonstrate the effectiveness of our baselines. Code is made available at https://github.com/JHome1/GOSS_Segmentor. [ABSTRACT FROM AUTHOR]

Cryptography is required while interacting through an insecure media, such as internet. Unauthorised individuals have the ability to view and alter data as it is being transferred from one location to other. Cryptography is important in the realm of security of data. Cryptographic methods are being utilized to protect information against attackers in various areas. Symmetric key encryption employs a single key for encryption and decryption, and is a famously used cryptosystem. In this manuscript, improved elliptical curve cryptography and chaotic mapping with fruitfly optimization algorithm is proposed for Secure Data Transmission (IECC-CMFOA-SDT). Here, cryptography and steganography are consolidated to enhance the data security. Initially, input is the plain text from the sender and compression is done to decrease the size of input data. Then data obfuscation is carried out over the compressed data file with the aid of binary conversion, 2's complement, decimal conversion, and ASCII conversion. By utilizing the improved elliptical curve cryptography (IECC) algorithm, obfuscated data is encrypted. Encrypted data is embedded with the support of Chaotic mapping based fruitfly optimization algorithm (CMFOA), which conceals the data by jumbling the pixels in the image. Hence the secret data is diversified. CMFOA picks the superior values of pixels to root the encrypted data. The receiver acquires the stego image with embedded concealed data, which chooses the optimum pixels points to extract the concealed data from the image. Original data is acquired by decrypting, deobfuscating and decompressing the extracted data. The proposed approach is executed in CloudSim Plus. The performance metrics, like obfuscation time, deobfuscation time, encryption time, decryption time, MSE and PSNR are analysed. Performance of IECC-CMFOA-SDT model gives 39.01%, 28.34%, and 37.45% lower encryption time and 17.12%, 24.12% and 32.07 higher PSNR compared with the existing methods, like optimal users based secure data transmission with lightweight block ciphers (CRT-HTLBO-SDT), Secure data transmission employing a hybrid of a bit mask oriented genetic algorithm, encryption and steganography (BMOGA-DWT-SDT) and cryptanalysis of an encryption method based on a compound coupled logistic map and an anti-codifying technique for secure data transmission (CCLM-ACT-SDT). [ABSTRACT FROM AUTHOR]

Chest X-ray (CXR) images serve as a fundamental diagnostic tool in the field of medical imaging. Accurate and robust lung segmentation in CXR images is a crucial step toward prediction of age, automating disease diagnosis and monitoring. Deep learning algorithms have been fig used in the segmentation process. The low-resolution output closely resembles the original high-resolution image. However, after being resampled, the image borders become blurred and, in some cases, CXR images may have low contrast between lung tissue and surrounding structures, making it challenging for algorithms to distinguish between them accurately. To overcome the difficulties noted above, a novel approach is introduced for lung segmentation in CXR images, which combines the power of CXR-Autoencoder Segmentation (CXR-AES) with Markov Random Fields (MRF) to achieve enhanced precision and performance. The CXR-AES component is responsible for feature extraction and initial segmentation, while the MRF serves as a contextual model that refines the segmentation results by considering spatial dependencies among pixels. This synergistic fusion of techniques enables the model to capture intricate lung boundaries and handle challenging cases, including pathologies and image artifacts. Finally, the above method gives a good result with dice, sensitivity, specificity, and precision performance metrics of 95.6, 89.9, 99.5, and 91.0% on segmented masks and lung, respectively. CXR-AES with MRF in lung segmentation has broad implications for clinical practice, research, and healthcare innovation. It enhances the efficiency of lung region extraction in CXR images, ultimately improving the diagnosis, treatment, and management of lung-related conditions. [ABSTRACT FROM AUTHOR]

The complex world of cellular dynamics has never been easier to understand, because of the developments in Bio imaging technology. When it comes to precisely investigating subcellular structures, traditional imaging techniques are useful but sometimes ithave drawbacks. In this work, we provide a novel Quantum Bioimaging Paradigm (QBP) intended to resolve subcellular localization anomalies at a resolution until thought to be impossible.Subcellular abnormalities in complex biological structures are disgracefully difficult for conventional tools to capitalize on properly. A novel approach is called Improved Quantum Convolutional Neural Network (IQ-CNN), which is important to use feature selection methods to ensure that the features are optimal. Anomaly detection relies on identifying and retaining important characteristics. Collecting the dataset contains Human Protein Atlas (HPA), and immunohistochemistry (IHC) microscopy images for training and validation. We employ Decimal Scaling Normalization to standardize pixel values, ensuring consistent representation across diverse Bioimaging datasets. Extract the feature from the normalized Bioimaging dataset using the Histogram of Oriented Gradients (HOG) method. Our investigations assess the suggested model's efficacy by contrasting the IQ-CNN with other methods. Accuracy (97%), precision (87%), recall (82%), and F1-score (95%), among other assessments, are used. The suggested method shows excellent results in terms of enhanced subcellular anomaly detection reliability. These outcomes are improvement advance for quantum-enhanced Bioimaging, a rapidly developing area with promising potential for use in medical diagnostics and treatment. [ABSTRACT FROM AUTHOR]

We propose a linear image sensor specialized for the light section method to achieve high-speed and low-latency height measurement. This linear image sensor has pixels in the shape of triangles pointing upward and downward, which allows fast data readout and simple data processing of the cross-sectional profile acquisition. We have confirmed the basic feasibility of the sensor by numerical simulation. We also fabricated a prototype of the proposed sensor with 508 pixels. We conducted height measurement of the vertical translate stage and the step pyramid-shaped object and confirmed that the proposed sensor functions as a sensor for light section method. [ABSTRACT FROM AUTHOR]

This paper proposes a new and improved one-dimensional chaotic system (1DL-SCS). After performance testing, it is verified to have better chaotic performance and a larger range of parameters. Based on this, a new image encryption algorithm is proposed. First, the chaotic sequences generated by the hyper-chaotic Chen system are used to perform twice circular rotations for achieving efficient global scrambling. Then we index scramble the image by row using 1DL-SCS and redistribute the pixel points by column, which further breaks the correlation between image pixels and prepares for diffusion. In the diffusion stage, the plaintext is divided into four parts for processing with the improved generalized Feistel structure, and the plaintext sequence in the round function is further divided into two parts, which greatly improves the diffusion efficiency. Through two rounds of iterative operations, a good diffusion effect can be achieved efficiently. Simulation experiment and performance analysis indicates that the algorithm has outstanding advantages such as excellent ciphertext statistical characteristics, and strong resistance to chosen plaintext attacks. [ABSTRACT FROM AUTHOR]

Due to the proliferation of high-quality copying devices and the significant profits of counterfeit products, it is critical to establish an effective scheme for detecting and preventing the counterfeiting of goods. At present, most anti-faking schemes leave much to be desired in terms of cost, convenience, and ability to facilitate pre-sale authentication. The paper designs a unique textured pattern and proposes a triple anti-counterfeiting authentication (TACA). First, the textured pattern consists of triple encryptions (the first is that the key area of the QR code is covered, the second includes scale and Arnold transformation, and the third involves replacing the black areas of the pattern with random multi-level grayscales), the abundant details in the texture not only effectively conceal information, but also that their distortion will increase. Second, TACA comprises interpretability analysis (IA), spectral feature analysis (SFA), and spot matching analysis (SMA) in a cascaded way. In further detail, IA mainly exploits the positional transformation of individual pixels and the block features of local regions to restore interpretability. SFA uses the low-frequency subgraph of discrete wavelet transform (DWT) at a specified scale to capture macroscopic structural information. SMA is able to capture the detailed information of the pattern by utilizing SURF to detect the peak region rate positions and employing BRISK to accurately describe them before. Finally, this paper investigates the robustness of the proposed anti-counterfeiting scheme under a variety of copying methods (replicating, scanning-printing), capturing devices (smartphones), and attack scenarios (no attack, cropping, noise, blur). [ABSTRACT FROM AUTHOR]

Clothing image segmentation is a method to predict the clothing category label of each pixel in the input image. We reduced the influence of the variability of image shots, the similarity of clothing categories, and the complexity of boundaries on the segmentation accuracy of clothing images by developing an advanced ResNet50-based semantic segmentation model in this study whose primary structure is the encoder–decoder. An improved spatial pyramid pooling module combined with a global feature extraction branch of a large convolution kernel is developed to achieve multi-scale feature fusion and improve the model's ability to identify clothing and its boundary features in different shots. Furthermore, to balance the clothing shape and category information in the model, a spatial and semantic information enhancement module is proposed, which can enhance the circulation of the information between different stages of the network through cross-stage connection technology. The model was finally trained and tested on the Deepfashion2 dataset. The comparison experiment demonstrates that the proposed model obtained the highest mIoU and Boundary IoU of 74.55% and 57.51%, respectively, compared with the DeepLabv3+, PSPNet, and other networks. [ABSTRACT FROM AUTHOR]

The strength of the bond between concretes is attributed to surface roughness. Therefore, it is necessary to quantify the roughness of concrete surfaces. In the traditional method, concrete surface roughness is measured using the sand filling method, which is contacting, inefficient, and destructive. Accordingly, this paper proposes an edge detection-based method for measuring concrete surface roughness. By combining pixel and sub-pixel edge detection, aggregate edges are extracted and a relationship curve between edge frequency and concrete surface roughness is fitted using the least squares method to achieve a non-contact, stable and non-destructive measurement. Experimental outcomes indicate that the accuracy of the method for measuring concrete surface roughness can reach 94.7%. Additionally, a single-input, single-output neural network model based on calculated edge frequencies is constructed with a measurement accuracy of 95.4% for concrete surface roughness detection. [ABSTRACT FROM AUTHOR]

Reproducing correct soft shadows in interactive applications is accomplished by gathering multiple samples of the area light source. The size of the penumbra region is unpredictable, so the required sample density should be high to ensure soft shadow transitions in all configurations of the area light size. Therefore, multiple shadow maps must be evaluated for each scene configuration, which leads to performance degradation. To tackle the challenge, we propose a novel technique based on the multisampling depth, which approximates soft shadows for large, fully dynamic scenes at interactive frame rates. Our algorithm can perform multiple depth tests for each pixel in the shadow map. We effectively reduce the number of samples of zone light to avoid other rendering passes. We reuse shadow maps for other light source sampling setups without performing additional scene rendering while maintaining real-time performance. Our approach fits well in existing rendering pipelines and improves the robustness of many different soft shadow techniques. [ABSTRACT FROM AUTHOR]

Quantum medical image encryption is one of the new methods to send medical images in an insecure communication channel. The proposed method is based on the flexible representation of quantum images which consists of two phases. In the first phase, the medical image is first interlaced using C-NOT gates and then the output of this phase is given to the second phase. In the second phase, one key is defined based on the logical mapping chaos function and the medical image encryption operation is performed based on this key. The output of the second phase is an encrypted image that can only be decrypted by the receiver which has the key. The results of the study of criteria such as histogram, entropy, correlation coefficient of adjacent pixels and correlation diagram show good performance. The entropy value of the proposed method is 99.67 which shows the proposed algorithm has much better performance than previous work. The temporal complexity of the designed quantum circuits is O(n), while the temporal complexity of the previous work is equal to O 2 2 n , which indicates the fast implementation of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method. [ABSTRACT FROM AUTHOR]

Aiming at the problems of identifying storage pest tobacco pest beetles from images that have few object pixels and considerable image noise, and therefore suffer from lack of information and identifiable features, this paper proposes an automatic monitoring method of tobacco beetle based on Multi-scale Global residual Feature Pyramid Network and Dual-path Deformable Attention (MGrFPN-DDrGAM). Firstly, a Multi-scale Global residual Feature Pyramid Network (MGrFPN) is constructed to obtain rich high-level semantic features and more complete information on low-level features to reduce missed detection; Then, a Dual-path Deformable receptive field Guided Attention Module (DDrGAM) is designed to establish long-range channel dependence, guide the effective fusion of features and improve the localization accuracy of tobacco beetles by fitting the spatial geometric deformation features of and capturing the spatial information of feature maps with different scales to enrich the feature information in the channel and spatial. Finally, to simulate a real scene, a multi-scene tobacco beetle dataset is created. The dataset includes 28,080 images and manually labeled tobacco beetle objects. The experimental results show that under the framework of the Faster R-CNN algorithm, the detection precision and recall rate of this method can reach 91.4% and 98.4% when the intersection ratio (IoU) is 0.5. Compared with Faster R-CNN and FPN, when the intersection ratio (IoU) is 0.7, the detection precision is improved by 32.9% and 6.9%, respectively. The proposed method is superior to the current mainstream methods. [ABSTRACT FROM AUTHOR]

Accurate labeling of lung nodules in computed tomography (CT) images is crucial in early lung cancer diagnosis and before nodule resection surgery. However, the irregular shape of lung nodules in CT images and the complex lung environment make it much more challenging to segment lung nodules accurately. On this basis, we propose an improved V-Net segmentation method based on pixel threshold separation and attention mechanism for lung nodules. This method first offers a data augment strategy to solve the problem of insufficient samples in 3D medical datasets. In addition, we integrate the feature extraction module based on pixel threshold separation into the model to enhance the feature extraction ability under different thresholds on the one hand. On the other hand, the model introduces channel and spatial attention modules to make the model pay more attention to important semantic information and improve its generalization ability and accuracy. Experiments show that the Dice similarity coefficients of the improved model on the public datasets LUNA16 and LNDb are 94.9% and 81.1% respectively, and the sensitivities reach 92.7% and 76.9% respectively. which is superior to most existing UNet architecture models and comparable to the manual level segmentation results by medical technologists. [ABSTRACT FROM AUTHOR]

This paper addresses few-shot semantic segmentation and proposes a novel transductive end-to-end method that overcomes three key problems affecting performance. First, we present a novel ensemble of visual features learned from pretrained classification and semantic segmentation networks with the same architecture. Our approach leverages the varying discriminative power of these networks, resulting in rich and diverse visual features that are more informative than a pretrained classification backbone that is not optimized for dense pixel-wise classification tasks used in most state-of-the-art methods. Secondly, the pretrained semantic segmentation network serves as a base class extractor, which effectively mitigates false positives that occur during inference time and are caused by base objects other than the object of interest. Thirdly, a two-step segmentation approach using transductive meta-learning is presented to address the episodes with poor similarity between the support and query images. The proposed transductive meta-learning method addresses the prediction by first learning the relationship between labeled and unlabeled data points with matching support foreground to query features (intra-class similarity) and then applying this knowledge to predict on the unlabeled query image (intra-object similarity), which simultaneously learns propagation and false positive suppression. To evaluate our method, we performed experiments on benchmark datasets, and the results demonstrate significant improvement with minimal trainable parameters of 2.98M. Specifically, using Resnet-101, we achieve state-of-the-art performance for both 1-shot and 5-shot Pascal- 5 i , as well as for 1-shot and 5-shot COCO- 20 i . [ABSTRACT FROM AUTHOR]

Identifying and extracting urban features such as buildings and producing accurate information about the location of features from remotely sensed images is important in updating maps, spatial databases, urban planning, and meeting the needs of urban services. Supervised methods of producing training data were widely applied in deriving accurate information from such images. However, those approaches require human intervention and are therefore time-consuming, and associated with classification accuracy limitations. This study presents a novel approach for extracting buildings from remotely sensed images by automating training data selection steps for supervised classification. To prepare training samples intelligently and automatically, a knowledge-based integration of radar and optical images was used. First, a general height constraint was applied to a digital surface model (DSM) obtained from radar images to divide the image pixels into two parts. Then, by applying a local height filter in each part, pixels representing buildings were extracted. Considering that the buildings are not the only features with significant heights, another constraint defined as coherence constraint was considered. Then, histogram of the optical image pixels corresponding to the filtered radar DSM pixels extracted from the previous steps was used to select sample training pixels. By considering pixels within the maximum range of the histogram, the final building samples for Support Vector Machine (SVM) classification were derived and applied for building extraction. The output showed that it sufficiently and reliably improved the existing supervised classification limitation to extract building. The performance of the proposed method was evaluated over different image datasets regardless of input image type and region. It achieved an overall average accuracy of 92% and a kappa coefficient of 0.83. [ABSTRACT FROM AUTHOR]

Images taken under the low-light condition often contain blur and saturated pixels at the same time. Deblurring images with saturated pixels is quite challenging. Because of the limited dynamic range, the saturated pixels are usually clipped in the imaging process and thus cannot be modeled by the linear blur model. Previous methods use manually designed smooth functions to approximate the clipping procedure. Their deblurring processes often require empirically defined parameters, which may not be the optimal choices for different images. In this paper, we develop a data-driven approach to model the saturated pixels by a learned latent map. Based on the new model, the non-blind deblurring task can be formulated into a maximum a posterior problem, which can be effectively solved by iteratively computing the latent map and the latent image. Specifically, the latent map is computed by learning from a map estimation network, and the latent image estimation process is implemented by a Richardson–Lucy (RL)-based updating scheme. To estimate high-quality deblurred images without amplified artifacts, we develop a prior estimation network to obtain prior information, which is further integrated into the RL scheme. Experimental results demonstrate that the proposed method performs favorably against state-of-the-art algorithms both quantitatively and qualitatively on synthetic and real-world images. [ABSTRACT FROM AUTHOR]

Color constancy is still one of the biggest challenges in camera color processing. Convolutional neural networks have been able to improve the situation but there are still problems in many conditions, especially in scenes where a single color is dominating. In this work, we approach the problem from a slightly different setting. What if we could have some other information than the raw RGB image data. What kind of information would help to bring significant improvements while still be feasible in a mobile device. These questions sparked an idea for a novel approach for computational color constancy. Instead of raw RGB images used by the existing algorithms to estimate the scene white points, our approach is based on the scene's average color spectra-single pixel spectral measurement. We show that as few as 10–14 spectral channels are sufficient. Notably, the sensor output has five orders of magnitude less data than in raw RGB images of a 10MPix camera. The spectral sensor captures the "spectral fingerprints" of different light sources and the illuminant white point can be accurately estimated by a standard regressor. The regressor can be trained with generated measurements using the existing RGB color constancy datasets. For this purpose, we propose a spectral data generation pipeline that can be used if the dataset camera model is known and thus its spectral characterization can be obtained. To verify the results with real data, we collected a real spectral dataset with a commercial spectrometer. On all datasets the proposed Single Pixel Spectral Color Constancy obtains the highest accuracy in the both single and cross-dataset experiments. The method is particularly effective for the difficult scenes for which the average improvements are 40–70% compared to state-of-the-arts. The approach can be extended to multi-illuminant case for which the experimental results also provide promising results. [ABSTRACT FROM AUTHOR]

Local neighborhood information plays an important role in robust fuzzy clustering-related segmentation algorithms, and how to construct local information items is the key to robust fuzzy clustering. Based on existing local information constraints, this paper proposes a model to describe the hierarchy relationship of local neighborhood windows in the master–slave neighborhood model, which combines spatial distance with gray information to suppress the influence of noise on current pixel clustering, and can also well control the balance of noise suppression and detail preservation. Based on this model, this paper proposes a robust fuzzy clustering segmentation algorithm with master-slave neighborhood information constraints. When constraining the neighborhood pixels of a pixel (that is the master neighborhood pixels), the algorithm will further constrain the pixels in the neighborhood window around the master neighborhood pixel (that is the salve neighborhood pixels), thus enhancing the robustness of the algorithm. Experiments results show that the proposed algorithm has good segmentation performance and strong anti-noise performance, even significantly outperforms existing state-of-the-art robust fuzzy clustering-related algorithms in the presence of high noise. [ABSTRACT FROM AUTHOR]

This paper presents an image encryption scheme with data and appearance security, by adopting inter-intra-block scrambling and weighted diffusion. The 2D robust hyper-chaotic map with flexible geometric distribution and rich hyper-chaotic parameter space is employed to generate the key stream for encryption, by considering the characteristics of plaintext. The plain image is first preprocessed by Huffman coding for getting compressed image. Then, the compressed image is divided into four sub-blocks and is further permuted by the designed inter-intra-block scrambling scheme, which can improve the scrambling effect by making the pixel far away from the original adjacent pixels. After that, a weighted diffusion method strongly related to plaintext and key stream is introduced to diffuse the shuffled image to obtain the noise-like cipher image. And in pursuit of higher security, the meaningless noise-like image is embedded into host image to create the visually meaningful cipher image. A series of experiment tests and analyses are carried out to further demonstrate the excellent performances of the encryption scheme. [ABSTRACT FROM AUTHOR]

The highly sensitive millimeter-wave telescope is an important tool for accurate measurement of cosmic microwave background (CMB) radiation, and its core component is a detector array located in a cryogenic focal plane. The feasibility of utilizing thermal kinetic inductance detectors (TKIDs) for CMB observations has been demonstrated. We propose a pixel design of TKIDs for observing CMB through atmospheric windows for observations in the 90/150 GHz bands. The TKIDs are designed to achieve photon noise limited sensitivity with overall noise equivalent power can be less than 20 aW/ Hz . Silicon-rich silicon nitride is used as the dielectric of TKIDs, and Al is used as the inductance material. Two pairs of probes are designed on the pixel to divide the signal into two polarization directions. Orthogonal transducer and diplexer are used for signal conversion and frequency division. Assuming lossless dielectric, the coupling efficiency of a single pixel is around 90%. This pixel design will be utilized for future large-scale TKIDs array designs for CMB observations. [ABSTRACT FROM AUTHOR]

SpectroPhotometric Imaging in Astronomy with Kinetic Inductance Detectors (SPIAKID) aims at designing, building, and deploying on the sky a spectrophotometric imager based on microwave kinetic inductance detectors (MKIDs) in the optical and near-infrared bands. MKIDs show a fast response and the ability to resolve photon energy compared to the conventional Charge-coupled Devices (CCDs). In this paper, we present the design and simulation of the MKID arrays for SPIAKID. The detectors consist of four arrays with each array of 20,000 lumped-element pixels, and each array will be read with 10 readout lines. The meander material of the resonators is trilayer TiN/Ti/TiN to have better uniformity of the critical temperature across the array. We also present the measurement result for a test array with 30 × 30 pixels which is a subset of the designed 2000-pixel array to verify the design and fabrication. The current measured best energy-resolving power R = E / Δ E is 2.4 at λ = 405 nm and the current medium R is around 1.7. We have also observed that the response of the TiN/Ti/TiN is much smaller than expected. [ABSTRACT FROM AUTHOR]

Gesture recognition using cameras capable of capturing detailed images for gesture recognition is not feasible in many places due to concerns regarding privacy and information leakage. To address this problem, we have proposed a method of capturing shadow pictures using single-pixel-imaging to realize privacy-conscious gesture recognition. As an implementation method of single-pixel-imaging in public spaces, we have studied using a high-frame-rate LED display as a light source. By using a high-frame-rate LED display, random patterns can be latent while the observer perceives an apparent image. However, the image reconstructed by single-pixel-imaging using a high-frame-rate LED display is influenced by the apparent image, making gesture recognition difficult. In this study, we show that the influence of the apparent image can be removed by restoring the restored image using deep learning with a convolutional network called U-Net, and high classification accuracy with a small number of illuminations by using LeNet to classify restored images. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have been widely used in image classification tasks and have achieved remarkable results compared with traditional methods. Their main advantage is the ability to extract hidden features automatically using local connectivity and spatial locality. However, CNN cannot be applied to tabular data, mainly due to the unsuitability of the tabular data structure to the CNN input. In this paper, we propose a new generic method for the representation of multidimensional tabular data as color-encoded images that can be used both for data visualization and classification with CNN. Our approach, named FC-Viz (Feature Clustering-Visualization), is based on user-oriented data visualization ideas, such as pixel-oriented techniques, feature clustering, and feature interactions. The proposed approach includes a transformation of each instance of the tabular data into a 2D pixel-based representation, where pixels representing features with strong correlation and interaction are adjacent to each other. We applied FC-Viz to ten multidimensional tabular datasets with dozens to thousands of features and compared its classification and visualization performance with a state-of-the-art tabular data transformation method. The evaluation experiments show that our approach is as accurate as the state-of-the-art, but with much smaller images resulting in much more compact and faster CNN models. [ABSTRACT FROM AUTHOR]

With the rapid development of multimedia technology and the massive accumulation of user data, a huge amount of data is rapidly generated and shared over the network, while the problems of inappropriate data access and abuse persist. Reversible data hiding in encrypted images (RDHEI) is a privacy-preserving method that embeds protected data in an encrypted domain and accurately extracts the embedded data without affecting the original content. However, the amount of embedded data has been one of the major limitations in the performance and application of RDHEI. Currently, the main approaches to improve the capacity of RDHEI are either to increase the overall capacity or to reduce the length of the auxiliary information. In this paper, we propose a novel RDHEI scheme based on multi-prediction and adaptive Huffman encoding. To increase the overall capacity, we propose a multi-prediction, called MED+GAP predictor, to generate the label map data of non-reference pixels prior to image encryption. Then, an adaptive Huffman coding is designed to compress the generated labels in order to reduce the embedding length of the auxiliary information used for the extraction and recovery. Experiments show that the proposed method with MED+GAP predictor and adaptive Huffman coding improves 0.052 bpp, 0.023 bpp, and 0.047 bpp on average over the other state-of-the-art methods on the BOSSBase, BOWS-2, and UCID datasets, respectively, while maintaining security and reversibility. [ABSTRACT FROM AUTHOR]

Recovery or filling in of missing pixels in damaged images is a challenging problem known as image inpainting. Many currently used techniques still suffer from artifacts and other visual defects. In the proposed inpainting approach, the authors have combined wavelet patch-based and group-based sparse representation learning so as to exploit the benefits of (a) multiresolution decomposition using wavelets, (b) sparsity and (c) coherence. The proposed method creates multiple dictionaries employing adaptive K-SVD (K-singular value decomposition) on wavelet decomposed components. The method also creates another dictionary employing PCA (principal component analysis) on group-based image patches. Finally, to accomplish the operation of inpainting, dictionaries of both types are integrated using the ADMM (alternating direction method of multipliers). The proposed method has been tested on images with varying degrees of degradation in terms of the percentage of missing pixels or blocks. We have rated the performance and compared proposed method with other state-of-the-art inpainting methods based on measures like the peak signal-to-noise ratio, the structural similarity index measure, and the figure of merit. The high values of the performance measures establish the efficacy of the proposed method. [ABSTRACT FROM AUTHOR]

Semantic segmentation suffers from boundary shift and shape deformation problems due to the neglect of overall guidance information. Motivated by that the object boundaries have a stronger representation for overall information of target objects, we propose a simple yet effective network, boundary-guidance network (BG-Net), for object consistency maintaining in semantic segmentation. In our work, we explore the pixels both near and far from the boundary, characterizing each pixel by utilizing pixel-boundary association. The boundary feature is integrated into the segmentation feature to mitigate boundary shift and shape deformation problem. We explicitly supervise the angle and distance information of pixels pointing to the nearest object boundary. Then the association can be learned by geometric modelling. Meanwhile, the low-level feature emphasized up-sampling (LFEU) module is designed to supplement the detail representation in high-level feature without direct interference. Finally, we evaluate our method on Cityscapes and CamVid datasets. The experimental results demonstrate the superiority of our BG-Net. [ABSTRACT FROM AUTHOR]

The aim of the reference-based image super-resolution (RefSR) is to reconstruct high-resolution (HR) when a reference (Ref) image with similar content as that of the low-resolution (LR) input is given. In the task, the quality of existing approaches degrades severely when there are several similar objects but different contents. Besides, not all similar information in the reference image is useful for the input image. Therefore, we propose high-to-low-level feature matching and complementary information fusion (HMCF) network for RefSR. The matching strategy adopts high-level to low-level feature matching to distinguish similar objects but different contents according to high-level semantics. The complementary information fusion module utilizes the channel and spatial attention to select the complement information for LR image and keeps the pixel consistency of input and Ref image. We perform extensive experiments to demonstrate that our proposed HMCF obtains the SOTA performance on the RefSR benchmarks and presents a high visual quality. [ABSTRACT FROM AUTHOR]

The application of continuous-time quantum walk in the field of image segmentation has attracted much attention due to the advantages of quantum computation. However, the proposed image segmentation algorithm constructs the continuous-time quantum walk model based on pixels, which will cause a huge burden on quantum resources, and the various feature information of the image cannot be better considered. In addition, this pixel-based processing method requires a lot of manual annotation to achieve the desirable segmentation effect. To address these issues, we propose an image segmentation algorithm using continuous-time quantum walk based on superpixels. In our segmentation algorithm, the original image is firstly segmented into superpixels, and then a weighted graph is constructed with superpixels as nodes, where the weight of edges in graph is measured by the feature similarity of two adjacent superpixels, which consists of color features and texture features. Next, the continuous-time quantum walk model is constructed based on the weighted graph by redefining the new Hamiltonian operator. Finally, continuous-time quantum walk is executed and the image segmentation result can be obtained, which is realized by assigning each superpixel the class label corresponding to the greatest probability. Experiments on the BSD500 dataset show that the proposed algorithm can significantly improve segmentation efficiency and accuracy while the manually selected seeds is reduced by 91%. More importantly, the new algorithm reduce the demision of the quantum walk system by more than 99%, which will yield a huge saving on the quantum resources. [ABSTRACT FROM AUTHOR]

This paper discusses the problem of classifying images of land impressions on discharged bullets in terms of the "match" and "non-match" categories. The research is aimed at improving the effectiveness of comparing land impression images by the congruent matching profile segments (CMPS) method. The scientific novelty of the approach is in supplementing the analysis with an additional independent feature, as well as in using the k-nearest neighbors algorithm at the final stage of trace comparison. The research shows that the accuracy of classification of the compared pairs of land impression images by the combined method is approximately 87%. The analysis by the CMPS method makes it possible to effectively compare land impression images with high resolution (approximately 1 μm per pixel). The research is of interest to developers of automated ballistic identification systems. [ABSTRACT FROM AUTHOR]

This work is devoted to improving the quality of segmentation of images of various scientific papers and legal acts by neural network models by training them using modified loss functions that take into account special features of images of the appropriate subject domain. The analysis of existing loss functions is carried out, and new functions are proposed that work both with the coordinates of bounding boxes and use information about the pixels of the input image. To assess the quality, a neural network segmentation model with modified loss functions is trained, and a theoretical assessment is carried out using a simulation experiment showing the convergence rate and segmentation error. As a result of the study, rapidly converging loss functions are created that improve the quality of document image segmentation using additional information about the input data. [ABSTRACT FROM AUTHOR]

High-resolution sensors onboard satellites are generally reputed for rapidly producing land-use/land-cover (LULC) maps with improved spatial detail. However, such maps are subject to uncertainties due to several factors, including the training sample size. We investigated the effects of different training sample sizes (from 1000 to 12,000 pixels) on LULC classification accuracy using the random forest (RF) classifier. Then, we analyzed classification uncertainties by determining the median and the interquartile range (IQR) of the overall accuracy (OA) values through repeated k-fold cross-validation. Results showed that increasing training pixels significantly improved OA while minimizing model uncertainty. Specifically, larger training samples, ranging from 9000 to 12,000 pixels, exhibited narrower IQRs than smaller samples (1000–2000 pixels). Furthermore, there was a significant variation (Chi2 = 85.073; df = 11; p < 0.001) and a significant trend (J-T = 4641, p < 0.001) in OA values across various training sample sizes. Although larger training samples generally yielded high accuracies, this trend was not always consistent, as the lowest accuracy did not necessarily correspond to the smallest training sample. Nevertheless, models using 9000–11,000 pixels were effective (OA > 96%) and provided an accurate visual representation of LULC. Our findings emphasize the importance of selecting an appropriate training sample size to reduce uncertainties in high-resolution LULC classification. [ABSTRACT FROM AUTHOR]

We present an application of strategy synthesis to enforce spatial properties. This is achieved by implementing a toolchain that enables the tools CATLib and VoxLogicA to interact in a fully automated way. The Contract Automata Library (CATLib) is aimed at both composition and strategy synthesis of games modelled in a dialect of finite state automata. The Voxel-based Logical Analyser (VoxLogicA) is a spatial model checker for the verification of properties expressed using the Spatial Logic of Closure Spaces on pixels of digital images. We provide examples of strategy synthesis on automata encoding motion of agents in spaces represented by images, as well as a proof-of-concept realistic example based on a case study from the railway domain. The strategies are synthesised with CATLib, while the properties to enforce are defined by means of spatial model checking of the images with VoxLogicA. The combination of spatial model checking with strategy synthesis provides a toolchain for checking and enforcing mobility properties in multi-agent systems in which location plays an important role, like in many collective adaptive systems. We discuss the toolchain's performance also considering several recent improvements. [ABSTRACT FROM AUTHOR]

An observation simulator is established for the Spectroscopy Focusing Array (SFA) and Polarimetry Focusing Array (PFA) onboard the planned enhanced X-ray Timing and Polarimetry observatory (eXTP). It consists of photon generation, imaging, detection, and event readout to generate data products, which can be analyzed by the standard astronomical analysis software. It is used to simulate a few astronomical sources to estimate and understand the impact of the payload and platform design configurations on the scientific goals of eXTP, including the background estimation of the central pixel in SFA, position dependence of silicon drift detector signals and its impacts, and pointing jitter requirements. The joint data analysis of the SFA and PFA payloads shows that the PFA image can help estimate the impact of nearby sources on the target source and select the proper pixels for the background estimation of the central pixel in SFA. The spectral and timing study of the millisecond pulsar depicts that the position dependence of silicon drift detector signals itself has an insignificant impact on the results. The type-C low-frequency quasi-periodic oscillations of black holes ranging from 0.01 Hz to 30 Hz are considered to raise the pointing jitter requirements of the telescope. In this case, the stability of the telescope should be less than 12 ″ to avoid any spurious modulation signal. These examples demonstrate the necessity of an end-to-end observation simulator in the space mission, which will be further tested and improved by the ground segment in a wider range of applications. [ABSTRACT FROM AUTHOR]

The diversity of fabric defects and the lack of defect samples make detecting fabric defects an important and challenging problem. Currently, unsupervised algorithms are widely used for surface defect detection as they do not require annotated data and therefore reduce the cost of data acquisition. This paper presents a multi-stage unsupervised fabric defect detection method based on DCGAN. The method consists of three stages: the GAN training, the encoder training, and the classifier training. The first two stages allow our model to reconstruct the test images. In the image reconstruction process, we use a linear weighted fusion method to reduce the interference of defects. When the reconstructed image is subtracted from the original, we get a residual map that highlights the defects. This pixel-level detection makes it easier to detect different types of defects. In addition, we introduce a classifier training phase to generate a likelihood map for the test images. Each pixel value in the likelihood map represents the probability of the original map having a defect in that location region. Finally, we fuse the residual map with the likelihood map and further perform threshold segmentation on the fused residual map. Our method uses a separate training strategy at each stage and learns from a set of image patches cropped out online. The experimental results are compared with other methods in recent years and validate the method's superiority in terms of f-score metrics. [ABSTRACT FROM AUTHOR]