Search

Memory is a fundamental hardware structure used in various electronic and multi-media products. The design and development of large-size memory architectures have become increasingly complex due to the growing demands of high-resolution image processing applications. This research work aimes to design a specialized memory architecture module for Sobel edge detection, an essential technique in image processing. The proposed architecture (memory pixel size) consists of memory unit, comprising rows and columns, is determined by the image resolution. For effective Sobel edge detection, the image pixels must be stored in memory, and read operations are performed to access a 3x3 matrix of nine pixels. A critical consideration in developing this memory architecture is power dissipation, which is mitigated through the application of Clock Gating techniques. The proposed pixel memory architecture is implemented using MATLAB and Xilinx ISE software with Verilog HDL. The image pixel memory is developed using Block RAMs (BRAMs) and registers, and the 3x3 pixel matrix required for the Sobel edge detector is generated. Simulation, synthesis, and power analysis are conducted for the image pixel memory across source images with various resolutions, including 10x40, 10x20, 128x128, 320x240, and 512x512. The results indicate reduced power dissipation from 30% to 40% due to Clock Gating. This work demonstrates the effectiveness of the proposed memory architecture in reducing power consumption while maintaining performance. Future work aims to further enhance the performance of image pixel memory by decreasing the number of registers and improving pixel access times. Such module will provide a more efficient and scalable solution for high-resolution image processing applications. [ABSTRACT FROM AUTHOR]

In the age of information technology, the secure transmission of sensitive data over public networks is essential. Steganography, a data-hiding technique, is used to transmit information covertly between parties. However, many existing algorithms face a significant challenge: balancing payload size with the quality of the stego image. This article presents a new steganographic framework, DualLSBStego, using Dual-LSB in spatial domain images to address this issue. The DualLSBStego adaptively selects the two least significant pixel bits to hide secret data, enhancing the tradeoff between capacity and image quality. Experimental results on standard datasets in steganography of digital images, SIPI images datasets, show that the DualLSBStego achieves a Peak Signal-to-Noise Ratio (PSNR) of up to 79.64 decibels (dB) with an Airplane cover image at a payload size of 1 kilobit (kb). The DualLSBStego model's performance was compared to recent Difference Expansion (DE) techniques, achieving a Structural Similarity Index Measure (SSIM) of 1, which indicates a high-quality stego image. [ABSTRACT FROM AUTHOR]

Introduction: Individual leaves in the image are partly veiled by other leaves, which create shadows on another leaf. To eliminate the interference of soil and leaf shadows on cotton spectra and create reliablemonitoring of cotton nitrogen content, one classification method to unmanned aerial vehicle (UAV) image pixels is proposed. Methods: In this work, green light (550 nm) is divided into 10 levels to limit soil and leaf shadows (LS) on cotton spectrum. How many shadow has an influence on cotton spectra may be determined by the strong correlation between the vegetation index (VI) and leaf nitrogen content (LNC). Several machine learning methods were utilized to predict LNC using less disturbed VI. R-Square (R²), root mean square error (RMSE), and mean absolute error (MAE) were used to evaluate the performance of the model. Results: (i) after the spectrum were preprocessed by gaussian filter (GF), SG smooth (SG), and combination of GF and SG (GF&SG), the significant relationship between VI and LNC was greatly improved, so the Standard deviation of datasets was also decreased greatly; (ii) the image pixels were classified twice sequentially. Following the first classification, the influence of soil on vegetation index (VI) decreased. Following secondary classification, the influence of soil and LS to VI can be minimized. The relationship between the VI and LNC had improved significantly; (iii) After classifying the image pixels, the VI of 2-3, 2-4, and 2-5 have a stronger relationship with LNC accordingly. Correlation coefficients ( ) can reach to 0.5. That optimizes monitoring performance when combined with GF&SG to predict LNC, support vector machine regression (SVMR) has the better performance, R², RMSE, and MAE up to 0.86, 1.01, and 0.71, respectively. The UAV image classification technique in this study can minimize the negative effects of soil and LS on cotton spectrum, allowing for efficient and timely predict LNC. [ABSTRACT FROM AUTHOR]

The article deals with the issues of hiding text information in a graphic file. Most often, one or two least significant bits of the image pixels are modified. To do this, each byte of the secret message is divided into 8 or 4 parts. The use of the least significant bits of the graphic file for transmitting a secret message significantly limits the size of the original message, in addition, it allows steganographic analysis programs to detect and decrypt the transmitted data. A formula for hiding textual information in image pixels is proposed. The algorithm for hiding information is that the bytes of the secret message are mixed with the bytes of pixels of the key image using a secret formula. The result is new bytes of image pixels. A steganography scheme has been developed for embedding secret text in random image pixels. Random bytes are pre-embedded in each pixel row of the original image. As a result of the operations, a key image is obtained. Text codes are embedded in random pixel bytes of a given RGB channel. To generate a secret message, the characters of the table of ASCII codes are used. The detection and decryption program compares the pixels of the received image with the pixels of the key image in the specified RGB channel and extracts the codes of the encrypted text. The use of abstract images as a key image significantly increases the reliability of the protection of confidential information, since in such images there is a random change in pixel values. Demonstration programs for encryption and decryption in the Python 3.5.2 programming language have been developed. A graphic file is used as the decryption key. The developed steganography scheme allows not only transmitting sensitive information, but also adding digital fingerprints or hidden tags to the image

Pulse coupled neural network (PCNN) is widely used in image processing because of its unique biological characteristics, which is suitable for image fusion. When combining PCNN with non‐subsampled contourlet (NSCT) model, it is applied in overcoming the difficulty of coefficients selection for subband of the NSCT model. However in the original model, only the grey values of image pixels are used as input, without considering that the subjective vision of human eyes lacks the sensitivity to the local factors of the image. In this study, the improved pulse‐coupled neural network model has replaced the grey‐scale value of the image and introduced the weighted product of the strength of the gradient of the image and the local phase coherence as the model input. Finally, compared with other multi‐scale decompositions‐based image fusion and other improved NSCT‐PCNN algorithms, the algorithm presented in this study outperforms them in terms of objective criteria and visual appearance.

Brain tumour identification with traditional magnetic resonance imaging (MRI) tends to be time-consuming and in most cases, reading of the resulting images by human agents is prone to error, making it desirable to use automated image segmentation. This is a multi-step process involving: (a) collecting data in the form of raw processed or raw images, (b) removing bias by using pre-processing, (c) processing the image and locating the brain tumour, and (d) showing the tumour affected areas on a computer screen or projector. Several systems have been proposed for medical image segmentation but have not been evaluated in the field. This may be due to ongoing issues of image clarity, grey and white matter present in a scan image, lack of knowledge of the end user and constraints arising from MRI imaging systems. This makes it imperative to develop a comprehensive technique for the accurate diagnosis of brain tumors in MRI images. In this paper, we introduce a taxonomy consisting of 'Data, Image segmentation processing, and View' (DIV) which are the major components required to develop a high-end system for brain tumour diagnosis based on deep learning neural networks. The DIV taxonomy is evaluated based on system completeness and acceptance. The utility of the DIV taxonomy is demonstrated by classifying 30 state-of-the-art publications in the domain of medFical image segmentation systems based on deep neural networks. The results demonstrate that few components of medical image segmentation systems have been validated although several have been evaluated by identifying role and efficiency of the components in this domain. [ABSTRACT FROM AUTHOR]

Current LiDAR-only 3D detection methods inevitably suffer from the sparsity of point clouds and insufficient semantic information. To alleviate this difficulty, recent proposals densify LiDAR points by depth completion and then perform feature fusion with image pixels at the data-level or result-level. However, these methods often suffer from poor fusion effects and insufficient use of image information for voxel feature-level fusion. Meanwhile, noises brought by inaccurate depth completion significantly degrade detection accuracy. In this paper, we propose PVConvNet, a unified framework for multi-modal feature fusion that cleverly combines LiDAR points, virtual points and image pixels. Firstly, we develop an efficient Pixel-Voxel Sparse Convolution (PVConv) to perform voxel-wise feature-level fusion of point clouds and images. Secondly, we design a Noise-Resistant Dilated Sparse Convolution (NRDConv) to encode the voxel features of virtual points, which effectively reduces the impact of noise. Finally, we propose a unified RoI pooling strategy, namely Multimodal Voxel-RoI Pooling, for improving proposal refinement accuracy. We evaluate PVConvNet on the widely used KITTI dataset and the more challenging nuScenes dataset. Experimental results show that our method outperforms state-of-the-art multi-modal based methods, achieving a moderate 3D AP of 86.92% on the KITTI test set. • A unified multi-modal fusion framework is proposed to combine image, virtual points and points to achieve more accurate 3D detection. • A pixel-voxel sparse convolution is designed to perform feature-level fusion of point clouds and images, effectively realizing voxel-based backbone fusion. • A noise-resistant dilated submanifold convolution is constructed to encode voxel features of virtual points, which significantly reduces the impact of noise. • The object occlusion rate is used in cross-modal GT-Sampling to balance near-distant virtual objects while ensuring the consistency of point clouds and images. • Extensive experiments on the popular KITTI and nuScenes datasets validate the superiority of PVConvNet, with competitive performance compared to state-of-the-art 3D detection methods. [ABSTRACT FROM AUTHOR]

When the image position measuring method is applied to mover position measurement of a linear motor, the selection of the target source image is crucial to measurement accuracy and anti‐interference capacity. In this study, the fence image is used as the original shooting surface, and different kinds of fence images are constructed according to their width standard deviation and grey gradient sum. First, the relationship between the fence image spectrum energy and the correlation of image pixels is analysed. Then the image spectrum energy is used to characterise the feature of fence images during screening to select an aperiodic fence image with a strong anti‐interference. Second, the phase singularity method is used to match the two images before and after displacement to obtain a sub‐pixel displacement value of two images. Finally, according to the system calibration coefficient, the actual displacement value of the linear motor mover is obtained. Experimental results show that the fence image screening from image spectrum energy exhibits more anti‐interference for the position measurement of the linear motor, and the measurement algorithm based on the phase singularity algorithm is more stable. The absolute error of displacement is <0.023 mm, and the relative error change within 0.054–0.862%. [ABSTRACT FROM AUTHOR]

In this study, the authors present a simple, reliable, fast, unrestricted‐shape geometry, and accurate algorithm which runs in O(log2n) time to find the axis‐parallel largest rectangle (LR) inside a given region of interest (ROI), where n is the image size in one dimension, which means that the proposed model can work in real time. The proposed approach is successful in detecting the LR of arbitrary orientation that is fully contained in the ROI as well. Also, the present algorithm can find the largest empty rectangle in a space containing a set of zero points, whether the axis‐parallel rectangle or the oriented one. The strategy followed here is to accelerate LR detection process by searching the rectangle with the largest area inscribed in the ROI, by starting first with the lowest‐resolution version of the original image for determining the LR four corners' coordinates, then next searching the new LR corners' scaled coordinates in the higher power resolutions in a multiple resolutions hierarchical model and therefore, a corresponding coarse‐to‐fine inference procedure recursively eliminates the search space of the LR four corners coordinates. For finding the largest oriented rectangle, the same hierarchical procedures are followed, but combined with rotation‐angle resolution. [ABSTRACT FROM AUTHOR]

Skin deformation is the foundation of defining appropriate ease distribution at different parts of the body for compression clothing. The purpose of this study is to introduce a novel method to determine skin deformation of lower limb in cycling. Ten male volunteers (height 175 ± 2.4 cm and BMI 22.7 ± 1.7 kg/m²) were selected as the subjects in this study. The skin areas from waist to knee of this subject's lower limb in 12 cycling postures were obtained with a gel rubbing method. Then image pixels were transplanted to calculate the deformation of these skin areas in different postures and the skin deformation in cycling was plotted with the interpolation function in MATLAB. The results showed that skin deformations on the areas of abdomen, groin, haunch bone, gluteal fold, knee, and peoples changed obviously in cycling. The skin deformation on the area of abdomen and peoples was about -30% and -50% separately. Whereas, the skin had an about 40% stretch deformation on the areas of haunch bone, gluteal fold and knee and at the position of inner thigh, buttock bumps, and the middle of front and posterior thigh, the skin deformation was about 15%. The study presents a reliable and convenient method to procure skin deformation for active body postures. The results can provide primary data for pattern designer to determine suitable ease distribution, reasonable divisions of clothing pattern and matching elastic fabric for each division of cycling shorts. [ABSTRACT FROM AUTHOR]

A fog level estimation method using intensity curves arranged with geometrical information is proposed. The curves, extracted from pixels in the road region and reflecting the tendency of intensity convergence under foggy conditions, are provided to the stacked auto‐encoders and the encoded features are classified into four fog levels by a neural network. Experimental results with clear and foggy road images of various places show that the intensity curve feature of the proposed method is effectively working not only in detecting the presence of fog, but also in estimating the fog level accurately. [ABSTRACT FROM AUTHOR]

This paper presents a new method for estimating nearshore wave height from a digital video sequence. The method identifies main wave breaking zones in the video records and estimates the height of breaking waves inside the detected breaking zones. A geometric rectification is applied to the resulting estimation to convert the height measurement from image pixels to meters. The validation of the algorithm was undertaken over three months at Surfers Paradise, Australia. The performance of the algorithm was demonstrated to be comparable with that of buoy-measured wave height, as well as manual estimates of the onshore wave height by a surf reporter. The results indicate that the method can be used as a cost-effective tool for long-term monitoring of nearshore wave conditions. [ABSTRACT FROM PUBLISHER]

Telah dikembangkan sebuah program pengolahan citra untuk radiografi digital yang disusun menggunakan perangkat lunak Borland C++ Builder 4 Professional. Untuk melakukan pengolahan terhadap citra radiografi digital,dengan pengolahan terhadap pixel-pixel dari citra. Teknik pengolahan citra, meliputi pengaturan kecerahan citra (image brightness), kontras citra (image contras), pelembutan citra (image smoothing), penajaman citra (image sharpening), deteksi tepi (edge detection), citra negatif (negative image), dan histogram. Hasil pengolahan citra dihasilkan citra radiografi digital yang mudah  diinterprestasi  lebih  teliti  oleh  pengamat.  Tujuan penelitian ini adalah mendesain perangkat lunak pengolahan citra radiografi digital. Perangkat yang telah disusun kemudian digunakan untuk melakukan pengolahan terhadap citra radiografi digital. Hasil penelitian menunjukkan adanya peningkatan kualitas citra radiografi digital, sehingga dapat membantu seorang radiographer dalam mendiagnosa atau menginterprestasikan citra radiografi digital lebih teliti. Citra radiografi awal dan citra hasilnya dapat dilihat pada layar monitor sehingga  kualitas  pengolahan  dapat dibandingkan secara langsung. The image processing program has been developed for digital radiography that is designed by using software Borland C++ Builder 4 Professional to process the radiography digital image with image pixels processing. Image processing technique consists of image brightness, image contrast, image smoothing, image sharpening, edge detection, and negative image and histogram. From the result of image processing it produced radiography image that is easier to interpret by the observer. The purpose of this research is to design the software of image processing for digital radiography. Then the software is used as the image processor for digital radiography. The result shows that there is a quality improvement of digital radiography image, so it can help the radiographer to diagnose or interpret the detail radiography image. We also can see the radiography image and its result in the monitor screen and compare the quality of processing directly.

Orthogonal subspace projection (OSP) is a powerful tool for dimensionality reduction (DR) in hyperspectral images (HSIs). In the OSP approach, the basis of the signal subspace must be estimated from the data themselves. Such estimation task is referred to as signal subspace identification (SSI). Most of the SSI methods in the literature are based on the analysis of the data second-order statistics (SOS) and have two main drawbacks: 1) They do not take into account the rare signal components (or rare vectors), 2) they assume that noise is spatially stationary. Rare vectors are those signal components that are present in pixels scarcely represented in the image and linearly independent on the signal components characterizing the rest of the image pixels. SOS-based SSI algorithms estimate the signal subspace addressing mostly the background and ignoring the presence of the rare pixels. This may be detrimental for the performance of detection algorithms when DR is adopted as a pre-processing step in small target detection applications. In this paper, a new technique for SSI in HSIs is presented. The algorithm is developed to account for both the abundant and the rare signal components. The method is derived by assuming a signal-dependent model for the noise affecting the data. This makes the SSI algorithm particularly suitable for the processing of images acquired by new generation sensors where, due to the improved sensitivity of the electronic components, noise includes a signal-dependent term. Results on simulated data are discussed, and the comparison with a recently proposed technique based on the analysis of SOS is performed. Furthermore, the results obtained by applying the SSI algorithm to a real HSI affected by signal-dependent noise are presented and discussed. [ABSTRACT FROM AUTHOR]

A novel illumination normalisation (IN) method using a convolutional neural network (CNN) is proposed. The proposed network is composed of the local pattern extraction (LPE) and illumination elimination (IE) layers. The LPE layers model the relationships between the pixels in each local region in order to handle various types of local shadow and shading in the face image. Based on the commonly used assumption about the illumination field, the IE layers generate illumination‐insensitive ratio images by calculating the ratio between the output pairs produced from the LPE layers. The final feature map obtained by combining the ratio images can possess an improved discriminative ability for face recognition (FR). For training the proposed network, the results produced by the Weber fraction‐based IN methods as ground truths are utilised. The experimental results demonstrate that the proposed network performs better in terms of FR accuracy compared with the conventional non‐CNN‐based method and it can be combined with any CNN‐based face classifier. [ABSTRACT FROM AUTHOR]

Timbers of 105 × 105 mm2 in cross section and 1000 mm long were computed tomography (CT) scanned with a commercial unit in equal time steps during a drying cycle. The study aimed to convert the scanning numbers into wood density values, observe the changes that occur during the dehydration process, and, from these, differentiate moisture content as a function of time and space. The reference for calculating the moisture content of each image pixel was the oven-dry wood scan. Analysis of the results allowed visualizing the water distribution, the mechanism of water removal, as well as check formation. Although only a single drying run was analyzed and a limited number of timbers were scanned, the results revealed trends and thus show promise, so further research is urged because it may lead to a better understanding of water movement and check formation in wood during kiln drying. [ABSTRACT FROM AUTHOR]

Depth from defocus (DFD) technique calculates the blur amount in images considering that the depth and defocus blur are related to each other. Existing DFD methods generally compute the blur at edge locations and solve an optimisation problem to propagate the blur from edges to all image pixels. Solving the pixel‐based optimisation problem is time‐consuming, posing the performance bottleneck. Moreover, the generated depth maps are not consistent in textured areas and the blur estimation may be incorrect in the regions with soft shadows. We address these problems by proposing a superpixel‐based blur estimation method. Experimental results show that the proposed method is not only faster than pixel‐based blur estimation, but also can improve depth data in textured regions and soft shadows. [ABSTRACT FROM AUTHOR]

Supervised subspace learning techniques have been extensively studied in biometrics literature; however, there is little work dedicated to: 1) how to automatically determine the subspace dimension in the context of supervised learning, and 2) how to explicitly guarantee the classification performance on a training set. In this paper, by following our previous work on unified subspace learning framework in our earlier work, we present a general framework, called parameter-free graph embedding (PFGE) to solve the above two problems by posing a general supervised subspace learning task as a semidefinite programming problem. The semipositive feature Gram matrix, namely the product of the transformation matrix and its transpose, is derived by optimizing a trace difference form of an objective function extended from that in our earlier work with the constraints that guarantee the class homogeneity within the neighborhood of each datum. Then, the subspace dimension and the feature weights are simultaneously obtained via the singular value decomposition of the feature Gram matrix. In addition, to alleviate the computational complexity, the Kronecker product approximation of the feature Gram matrix is proposed by taking advantage of the essential matrix form of image pixels. The experiments on simulated data and real-world data demonstrate the capability of the new PFGE framework in estimating the subspace dimension for supervised learning as well as the superiority in classification performance over traditional algorithms for subspace learning [ABSTRACT FROM PUBLISHER]

Robust signal hashing defines a feature vector that characterizes the signal, independently of "nonsignificant" distortions of its content. When dealing with images, the considered distortions are typically due to compression or small geometrical manipulations. In other words, robustness means that images that are visually indistinguishable should produce equal or similar hash values. To discriminate image contents, a hash function should produce distinct outputs for different images. Our paper first proposes a robust hashing algorithm for still images. It is based on radial projection of the image pixels and is denoted the Radial hASHing (RASH) algorithm. Experimental results provided on the USC-SIPI dataset reveal that the proposed RASH feature vector is more robust and provides much stronger discrimination than a conventional histogram-based feature vector. The RASH vector appears to be a good candidate to build indexing algorithms, copy-detection systems, or content-based authentication mechanisms. To take benefit from the RASH vector capabilities, video content is summarized into key frames, each of them characterizing a video shot and described by its RASH vector. The resulting video hashing system works in real time and supports most distortions due to common spatial and temporal video distortions. [ABSTRACT FROM PUBLISHER]

In many traffic‐related applications, such as traffic management and structural health monitoring for roads, an accurate estimation of a moving vehicle's size and shape is needed before proceeding further. However, due to the presence of cast shadows, these properties cannot be obtained accurately using common object detection systems. To deal with the problem of misclassifying shadows as foreground, various methods have been introduced. Most of these methods often fail to distinguish shadow points from the foreground object when the boundary between the umbra and the object is unclear due to camouflage. A novel method for detecting moving shadows of vehicles in real‐time applications is presented. The method is based on two measurements, namely, the illumination direction and the intensity measurements in the neighbouring pixels in a scanned line. A major advantage of using image lines for classification is the ability to solve the problem associated with camouflages. Experimental results show that the proposed method is efficient in real‐time performances and has achieved higher detection rate and discrimination rate when compared with two well‐known methods. [ABSTRACT FROM AUTHOR]

The raster scan method in visual information processing is the rectangular pattern of image capture and reconstruction which systematically covers the area progressively, one line at a time starting from the top left corner. This traditional scanning method might work well for still image processing but for moving images, a scanning method such as diagonal snake scan can provide a better object correlation in a sequence of image frames. A novel adaptive scanning method for object‐oriented visual information processing is proposed. Here, image pixels are collected starting from the centre of the image area and progressively extended outwards similar to the water drop ripple effect. Simulation results have shown that the proposed work uses up to 33% less processing time to cover the region‐of‐interest in an image compared to the raster scan method. Preliminary simulation results have also shown that with the proposed adaptive scanning methodology, there is a significant improvement in the performance of the object‐oriented visual compression. [ABSTRACT FROM AUTHOR]