Search

In this paper, we focus on two tasks: semantic segmentation and target detection in the visual semantic understanding of tennis sports images and we optimize the network structure to achieve a more complete location contour information mining of the target. In detail, we focus on a weakly supervised image semantic segmentation method based on null convolution pixel relations. To address the problem of incomplete pixel-level pseudo-labeling, we introduce a cavity convolution unit with multiple cavity rates and a self-attentive mechanism in the classification model to adaptively enhance the target regions and suppress other irrelevant regions while expanding the perceptual field to generate high-quality pixel-level pseudo-labeling and then train the semantic segmentation model. The final experimental results show that the hierarchical fusion algorithm proposed in this paper significantly outperforms other algorithms, and the overall classification accuracy of the tandem cavity neural network algorithm reaches 81% with good overall classification results. The recognition accuracy of static movements is higher than that of dynamic movements. [ABSTRACT FROM AUTHOR]

In recent years, with the development of remote sensing technology and the enhancement of the value of remote sensing images in military and civil fields, remote sensing image object segmentation has also received more and more attention. This paper mainly studies the application of instance segmentation based on deep convolutional neural network in the remote sensing image. This paper proposes an attention balanced feature pyramid module, which strengthens multi-level features and uses the attention module to suppress the interference features of noise in the complex background. In addiction, Soft-NMS is introduced to improve the performance of the network, and GIoU loss is introduced to improve the effect of object detection. The proposed network improves the average detection and segmentation accuracy (mAP) values from 4 1. 7 5 % and 3 5. 3 4 % to 4 3. 0 5 % and 3 6. 0 2 % , respectively. [ABSTRACT FROM AUTHOR]

Image semantic segmentation technology is one of the core research contents in the field of computer vision. With the improvement of computer performance and the continuous development of deep learning technology, researchers have more and more enthusiasm to study the actual effect and performance of image semantic segmentation. The results of deep semantic segmentation allow computers to have a more detailed and accurate understanding of images, and have a wide range of application needs in the fields of autonomous driving, intelligent security, medical imaging, remote sensing images, etc. However, the existing image semantic segmentation algorithms have the disadvantages of easy discontinuous results and insufficient prediction accuracy. In this paper, we take deep learning-based image semantic segmentation technology as the research object to explore the improvement of the image semantic segmentation algorithm and its application in road scenarios. First, this paper proposes MCU-Net method based on residual fusion and multi-scale contextual information. MCU-Net uses residual fusion module to deepen the network structure and improve the ability of U-Net to acquire deeper features. Then a top-down and bottom-up path is constructed for feature information between different levels, and the spatial and semantic information contained in shallow and deep features in the network is fully utilized by fusing features from different levels. In addition, an enhanced void space pyramid pooling module is added for feature information between the same levels, which enables the output features to have a larger range of semantic information. Second, this paper proposes the DAMCU-Net method based on attention mechanism and edge detection based on MCU-Net. DAMCU-Net extracts global contextual information by the attention mechanism optimization module, while fusing features using dense jump connections to facilitate the network to recover more spatial detail information during upsampling, and uses the FReLU activation function to improve the segmentation capability of the network for complex targets. For the edge information lost in the feature extraction process, the edge detection branch is added to supplement the feature information of the main path by feature fusion to achieve the optimization of the edge information. [ABSTRACT FROM AUTHOR]

The traditional embroidery identification technology cannot present the target image of the art network in a more comprehensive and three-dimensional manner. A research on the segmentation and synthesis of embroidery art images based on deep learning convolutional neural network is proposed. Based on the semantic image segmentation technology of deep learning, this paper analyzes the embroidery semantic image segmentation technology, obtains the information of image technology, analyzes the embroidery rendering technology of convolutional neural networks, and puts forward the embroidery rendering algorithm. In order to verify the effectiveness of the algorithm, a simulation test experiment was carried out on the target content image and the embroidery art network image. The test results show that compared with the traditional method, this method has more specific and flexible image generation, stronger three-dimensional sense, closer to the real art embroidery network, and the direction of its needlework is also more hierarchical. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks have shown advanced performance in accurately segmenting images. In this paper, an SNP-like convolutional neuron structure is introduced, abstracted from the nonlinear mechanism in nonlinear spiking neural P (NSNP) systems. Then, a U-shaped convolutional neural network named SNP-like parallel-convolutional network, or SPC-Net, is constructed for segmentation tasks. The dual-convolution concatenate (DCC) and dual-convolution addition (DCA) network blocks are designed, respectively, in the encoder and decoder stages. The two blocks employ parallel convolution with different kernel sizes to improve feature representation ability and make full use of spatial detail information. Meanwhile, different feature fusion strategies are used to fuse their features to achieve feature complementarity and augmentation. Furthermore, a dual-scale pooling (DSP) module in the bottleneck is designed to improve the feature extraction capability, which can extract multi-scale contextual information and reduce information loss while extracting salient features. The SPC-Net is applied in medical image segmentation tasks and is compared with several recent segmentation methods on the GlaS and CRAG datasets. The proposed SPC-Net achieves 90.77% DICE coefficient, 83.76% IoU score and 83.93% F1 score, 86.33% ObjDice coefficient, 135.60 Obj-Hausdorff distance, respectively. The experimental results show that the proposed model can achieve good segmentation performance. [ABSTRACT FROM AUTHOR]

Remote sensing techniques have been developed over the past decades to acquire data without being in contact of the target object or data source. Their application on land-cover image segmentation has attracted significant attention during recent years. With the help of satellites, scientists and researchers can collect and store high-resolution image data that can be further be processed, segmented, and classified. However, these research results have not yet been synthesized to provide coherent guidance on the effect of variant land-cover segmentation processes. In this paper, we present a novel model that augments segmentation using smaller networks to segment individual classes. The combined network is trained on the same data but with the masks, combined and trained using categorical cross entropy. Experimental results show that the proposed method produces the highest mean IoU (Intersection of Union) as compared against several existing state-of-the-art models on the DeepGlobe dataset. [ABSTRACT FROM AUTHOR]

This paper discusses how to efficiently recognize flowers based on a convolutional neural network (CNN) using multiple features. Our proposed work consists of three phases including segmentation by Otsu thresholding with particle swarm optimization algorithms, feature extraction of color, shape, texture and recognition with the LeNet-5 neural network. In the feature extraction, an improved H component with the definition of WGB value is applied to extract the color feature, and a new algorithm based on local binary pattern (LBP) is proposed to enhance the accuracy of texture extraction. Besides this, we replace ReLU with Mish as activation function in the network design, and therefore increase the accuracy by 8% accuracy according to our comparison. The Oxford-102 and Oxford-17 datasets are adopted for benchmarking. The experimental results show that the combination of color features and texture features generates the highest recognition accuracy as 92.56% on Oxford-102 and 93% on Oxford-17. [ABSTRACT FROM AUTHOR]

In recent years, deep learning-based networks have been able to achieve state-of-the-art performance in medical image segmentation. U-Net, one of the currently available networks, has proven to be effective when applied to the segmentation of medical images. A Convolutional Neural Network's (CNN) performance is heavily dependent on the network's architecture and associated parameters. There are many layers and parameters that need to be set up in order to manually create a CNN, making it a complex procedure. Designing a network is made more difficult by using a variety of connections to increase the network's complexity. Evolutionary computation can be used to set the parameters of CNN and/or organize the CNN layers as an optimization strategy. This paper proposes an automatic evolutionary method for detecting an optimal network topology and its parameters for the segmentation of clinical image using Grey Wolf Optimization algorithm. Also, Bi-Directional LSTM integrated in the skip connection to extract dense feature characteristics of image by combining feature maps extracted from encoded and previous decoded path in nonlinear way (MIS-GW-U-Net-BiDCLSTM) is proposed. The experimental results demonstrate that the proposed method attains 98.49% accuracy with minimal parameters, which is much better than that of the other methods. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (DCNNs) trained on the pixel-level annotated images have achieved improvements in semantic segmentation. Due to the high cost of labeling training data, their applications may have great limitation. However, weakly supervised segmentation approaches can significantly reduce human labeling efforts. In this paper, we introduce a new framework to generate high-quality initial pixel-level annotations. By using a hierarchical image segmentation algorithm to predict the boundary map, we select the optimal scale of high-quality hierarchies. In the initialization step, scribble annotations and the saliency map are combined to construct a graphic model over the optimal scale segmentation. By solving the minimal cut problem, it can spread information from scribbles to unmarked regions. In the training process, the segmentation network is trained by using the initial pixel-level annotations. To iteratively optimize the segmentation, we use a graphical model to refine segmentation masks and retrain the segmentation network to get more precise pixel-level annotations. The experimental results on Pascal VOC 2012 dataset demonstrate that the proposed framework outperforms most of weakly supervised semantic segmentation methods and achieves the state-of-the-art performance, which is 6 9. 3 % mIoU. [ABSTRACT FROM AUTHOR]

Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems, yet still faces many challenges. Building on convolutional neural networks (CNNs), medical image segmentation has achieved tremendous progress. However, owing to the locality of convolution operations, CNNs have the inherent limitation in learning global context. To address the limitation in building global context relationship from CNNs, we propose LGNet, a semantic segmentation network aiming to learn local and global features for fast and accurate medical image segmentation in this paper. Specifically, we employ a two-branch architecture consisting of convolution layers in one branch to learn local features and transformer layers in the other branch to learn global features. LGNet has two key insights: (1) We bridge two-branch to learn local and global features in an interactive way; (2) we present a novel multi-feature fusion model (MSFFM) to leverage the global contexture information from transformer and the local representational features from convolutions. Our method achieves state-of-the-art trade-off in terms of accuracy and efficiency on several medical image segmentation benchmarks including Synapse, ACDC and MOST. Specifically, LGNet achieves the state-of-the-art performance with Dice's indexes of 80.15% on Synapse, of 91.70% on ACDC, and of 95.56% on MOST. Meanwhile, the inference speed attains at 172 frames per second with 2 2 4 × 2 2 4 input resolution. The extensive experiments demonstrate the effectiveness of the proposed LGNet for fast and accurate for medical image segmentation. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have achieved prominent performance in a series of image processing problems. CNNs become the first choice for dense classification problems such as semantic segmentation. However, CNNs predict the class of each pixel independently in semantic segmentation tasks, spatial regularity of the segmented objects is still a problem for these methods. Especially when given few training data, CNN could not perform well in the details, isolated and scattered small regions often appear in all kinds of CNN segmentation results. In this paper, we propose a method to add spatial regularization to the segmented objects. In our method, the spatial regularization such as total variation (TV) can be easily integrated into CNN network and it produces smooth edges and eliminate isolated points. We apply our proposed method to Unet and Segnet, which are well-established CNNs for image segmentation, and test them on WBC and CamVid datasets, respectively. The results show that the details of predictions are well improved by regularized networks. [ABSTRACT FROM AUTHOR]

To design an efficient fusion scheme for the generation of a highly informative fused image by combining multiple images is still a challenging task in computer vision. A fast and effective image fusion scheme based on multi-resolution singular value decomposition (MR-SVD) with guided filter (GF) has been introduced in this paper. The proposed scheme decomposes an image of two-scale by MR-SVD into a lower approximate layer and a detailed layer containing the lower and higher variations of pixel intensity. It generates lower and details of left focused (LF) and right focused (RF) layers by applying the MR-SVD on each series of multi-focus images. GF is utilized to create a refined and smooth-textured weight fusion map by the weighted average approach on spatial features of the lower and detail layers of each image. A fused image of LF and RF has been achieved by the inverse MR-SVD. Finally, a deep convolutional autoencoder (CAE) has been applied to segment the fused results by generating the trained-patches mechanism. Comparing the results by state-of-the-art fusion and segmentation methods, we have illustrated that the proposed schemes provide superior fused and its segment results in terms of both qualitatively and quantitatively. [ABSTRACT FROM AUTHOR]

This paper proposes an incoming Deep Convolutional Neural Network (CNN) architecture for segmenting retinal blood vessels automatically from fundus images. Automatic segmentation performs a substantial role in computer-aided diagnosis of retinal diseases; it is of considerable significance as eye diseases as well as some other systemic diseases give rise to perceivable pathologic changes. Retinal blood vessel segmentation is challenging because of the excessive changes in the morphology of the vessels on a noisy background. Previous deep learning-based supervised methods suffer from the insufficient use of low-level features which is advantageous in semantic segmentation tasks. The proposed architecture makes use of both high-level features and low-level features to segment retinal blood vessels. The major contribution of the proposed architecture concentrates on two important factors; the first in its supplying of extremely modularized network architecture of aggregated residual connections which enable us to copy the learned layers from the shallower model and developing additional layers to identity mapping. The second is to improve the utilization of computing resources within the network. This is achieved through a skillfully crafted design that allows for increased depth and width of the network while maintaining the stability of its computational budget. Experimental results show the effectiveness of using aggregated residual connections in segmenting retinal vessels more accurately and clearly. Compared to the best existing methods, the proposed method outperformed other existing methods in different measures, comprised less false positives at fine vessels, and caressed more clear lines with sufficient details like the human annotator. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy is a predominant vision-threatening disease affecting working-aged people specifically. Timely diagnosis through early detection and prevention helps to reduce the risk of severe vision loss. Computer-aided diagnosis in retinal image analysis through Machine Learning techniques will help medical professionals perform their analysis better. Automated image processing through Convolutional Neural Networks has proven to be a promising technique, mainly in medical image segmentation. Convolutional Neural Network techniques like 3D CNN, Deep CNN and architectures like U-Net, V-Net, SegNet, and DeepMedic have outperformed medical image analysis results. However, the baseline CNN architectures struggle to retain high-quality information at the output and thus affect the performance. This work focuses on addressing the issue of translational variance and overfitting scenarios of U-Net architecture by experimenting with adaptable window sizes, pretrained weights and linear interpolation technique. A novel U-Net based architecture called RetU-Net that segments abnormal retinopathy lesion structures by retaining higher-level feature is proposed. The experiments are conducted using Indian Diabetic Retinopathy Image Dataset provided in "Diabetic Retinopathy: Segmentation and Grading Challenge" initially. The results have been compared with other state-of-art CNN architectures. Further evaluation is carried out on public datasets: STARE and DRIVE, and the performance is compared with U-Net based architectures used for retinal image segmentation. The proposed approach is efficient in terms of precision, sensitivity, specificity and accuracy. It received accuracy scores of 0.9612, 0.9712 while experimenting with STARE and DRIVE datasets respectively. [ABSTRACT FROM AUTHOR]

"Image-based" computational fluid dynamics (CFD) simulations provide insights into each patient's hemodynamic environment. However, the current standard procedures for creating CFD models start with manual segmentation and are time-consuming, hindering the clinical translation of image-based CFD simulations. This feasibility study adopts deep-learning-based image segmentation [hereafter referred to as Artificial Intelligence (AI) segmentation] to replace manual segmentation to accelerate the CFD model creation. Two published convolutional neural network-based AI methods (MIScnn and DeepMedic) were selected to perform CFD model extraction from three-dimensional (3D) rotational angiography data containing intracranial aneurysms. In this study, aneurysm morphological and hemodynamic results using the models generated by AI segmentation methods were compared with those obtained by two human users for the same data. Interclass coefficients (ICCs), Bland–Altman plots, and Pearson's correlation coefficients (PCCs) were combined to assess how well the AI-generated CFD models performed. We found that almost perfect agreement was obtained between the human and AI results for all 11 morphological parameters and five out of eight hemodynamic parameters, while a moderate agreement was obtained from the remaining three hemodynamic parameters. Given this level of agreement, using AI segmentation to create CFD models is feasible, given more developments. [ABSTRACT FROM AUTHOR]

Tumor computed tomography (CT) image segmentation can provide a basis for the determination of tumor location and type. Therefore, it is of great significance to study the automatic segmentation method of tumor CT images. To address the problem of poor segmentation effect of traditional automatic tumor CT images segmentation methods, we propose an automatic segmentation method for tumor CT images using deep convolutional neural networks (DCNNs). First, the CT tumor image is simplified. According to the features of the target region and the background region, the distribution features of the tumor region in the CT images are obtained by convolution calculation, and the feature extraction is completed by feature fusion. Second, based on the feature extraction results, a deep supervised network is constructed to determine the image depth, which lays a solid foundation for accurate segmentation of tumor regions. Finally, DCNN was used to construct automatic segmentation for tumor CT images, which achieves the automatic segmentation of tumor CT images by mode calculation. The results show that the segmented tumor region is close to the actual region and the maximum pixel loss coefficient is 0.07, the maximum segmentation sensitivity is 7865 kbps/s, the pixel segmentation specific coefficient and the segmentation edge distance are kept at a low level, which has a certain application value in the field of tumor CT images. [ABSTRACT FROM AUTHOR]

This study aimed to investigate the effect of an image denoising algorithm based on weighted low-rank matrix restoration on thyroid nodule ultrasound images. A total of 1000 original ultrasound image data sets of thyroid nodules were selected as the study samples. The nodule segmentation data set of thyroid ultrasound region of interest (ROI) images was drawn and acquired. By introducing multiscale features and an attention mechanism to optimize the U-Net model, an ultrasound image segmentation model (F-U-Net) was constructed. The performance of the traditional U network model and full convolutional neural network model (FCN) was analyzed and compared by simulation experiments. The results showed that the dice coefficient, accuracy, and recall of the improved loss function in this study were significantly higher than those of the traditional cross entropy loss function and dice coefficient loss function, and the differences were statistically significant (P < 0.05). The Dice coefficient, accuracy, and recall of the F-U-net model were significantly higher than those of the traditional FCN model and U-net model (P < 0.05). The diagnostic sensitivity, specificity, accuracy, and positive predictive value of the F-U-net model for benign and malignant thyroid nodules were significantly higher than those of the FCN model and U-net model (P < 0.05). In summary, the proposed F-U network can effectively process the ultrasound images of thyroid nodules, improve the image quality, and help to improve the diagnostic effect of benign and malignant thyroid nodules. It provides a data reference for segmentation and reconstruction of benign and malignant ultrasound images of thyroid nodules. [ABSTRACT FROM AUTHOR]

Synthetic Aperture Radar (SAR) represents a type of active remote sensing technology that uses microwave electromagnetic radiation to produce and send data to the surface of a target location. SAR imaging is frequently used in national security applications since it is unaffected by weather, geographical location, or time. In this system, many approaches are examined, to improve automation for segmentation and classification. The utilization of Deep Neural Networks (DNNs) to classify SAR images has gotten a lot of attention, and it usually requires several layers of deep models for feature learning. With insufficient training data, however, the DNN will get affected by the overfitting issue. The major purpose of this work is to make a development on introducing a new framework for SAR image segmentation and categorization using deep learning. Owing to the coherent nature of the backscattering signal, SARs create speckle noise in their images. If the image has noisy material, classification becomes more challenging. Hence, the pre-processing of the images is employed by linear spatial filtering to remove the noise. Further, the Optimized U-Net is used for the segmentation. For the segmented images, the Binary Robust Independent Elementary Features (BRIEF) concept is adopted as the feature descriptor. These features are inputted to the Convolutional Neural Network (CNN) with Tuned Weight DNN (C-TWDNN) for the classification. In both segmentation and classification, the parameter tuning is employed by the combination of Galactic Swarm Optimization (GSO) and Deer Hunting Optimization Algorithm (DHOA) called the Self-adaptive-Galactic Deer Hunting Optimization (SA-GDHO). Experiments are conducted on a variety of public datasets, demonstrating that our method is capable of outperforming various expert systems and deep structured architectures. [ABSTRACT FROM AUTHOR]

In standard U-net, researchers only use long skip connections to skip features from the encoding path to the decoding path in order to recover spatial information loss during downsampling. However, it would result in gradient vanishing and limit the depth of the network. To address this issue, we propose a novel deep fully residual convolutional neural network that combines the U-net with the ResNet for medical image segmentation. By applying short skip connections, this new extension of U-net decreases the amount of parameters compared to the standard U-net, although the depth of the layer is increased. We evaluate the performance of the proposed model and other state-of-the-art models on the Electron Microscopy (EM) images dataset and the Computed Tomography (CT) images dataset. The result shows that our model achieves competitive accuracy on the EM benchmark without any further post-process. Moreover, the performance of image segmentation on CT images of the lungs is improved in contrast to the standard U-net. [ABSTRACT FROM AUTHOR]