Search

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]

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]

In recent years, several studies have undergone automatic blood vessel segmentation based on unsupervised and supervised algorithms to reduce user interruption. Deep learning networks have been used to get highly accurate segmentation results. However, the incorrect segmentation of pathological information and low micro-vascular segmentation is considered the challenges present in the existing methods for segmenting the retinal blood vessel. It also affects different degrees of vessel thickness, contextual feature fusion in technique, and perception of details. A deep learning-aided method has been presented to address these challenges in this paper. In the first phase, the preprocessing is performed using the retinal fundus images employed by the black ring removal, LAB conversion, CLAHE-based contrast enhancement, and grayscale image. Thus, the blood vessel segmentation is performed by a new deep learning model termed optimized ResUNet + +. As an improvement to this deep learning architecture, the activation function is optimized by the J-AGSO algorithm. The objective function for the optimized ResUNet + + -based blood vessel segmentation is to minimize the binary cross-entropy loss function. Further, the post-processing of the images is carried out by median filtering and binary thresholding. By verifying the standard benchmark datasets, the proposed model outperforms and attains enhanced performance. [ABSTRACT FROM AUTHOR]

In this paper, a new detection method of pin defects based on image segmentation and ResNe-50 is proposed, which realizes the defect detection of faulty pins in many aviation connectors. In this paper, a new dataset image segmentation method is used to segment many aviation connectors in a single image to generate a dataset, which reduces the tedious work of manually labeling the dataset. In the defect detection model, based on ResNet-50, a ResNet-B residual structure is introduced to reduce the loss of features during information extraction; a continuously differentiable CELU is used as the activation function to reduce the neuron death problem of ReLU; a new deformable convolution network (DCN v2) is introduced as the convolution kernel structure of the model to improve the recognition of aviation connectors with prominent geometric deformation pin recognition. The improved model achieved 97.2% and 94.4% accuracy for skewed and missing pins, respectively, in the experiments. The detection accuracy improved by 1.91% to 96.62% compared to the conventional ResNet-50. Compared with the traditional model, the improved model has better generalization ability. [ABSTRACT FROM AUTHOR]

Industrial fume emissions are a major contributor to global warming, and accurate monitoring is necessary. However, current segmentation techniques for video monitoring of industrial fumes are limited by ineffective edge segmentation, lack of consideration of dynamic characteristics, and poor segmentation accuracy. To address these issues, a deep learning-based semantic video segmentation network is proposed in this paper. The network combines fume deformation information, employs LR-ASPP design for real-time performance, and spatio-temporal consistency to enhance semantic information in the dynamic region. A residual hybrid attention network is constructed for the motion region to minimize loss of motion information. The proposed network demonstrates strong anti-interference capability in complex environments, achieving over 10% improvement in IoU and F -score under a high-speed segmentation of 53 FPS. This technology can be integrated with automated systems to enable timely responses to hazardous situations, minimizing risks to workers and nearby communities. In summary, the proposed deep learning-based semantic video segmentation network has significant implications for improving environmental monitoring and management practices in industrial settings. [ABSTRACT FROM AUTHOR]

Recently, image segmentation based on graph cut methods has shown remarkable performance on a set of image data. Although the kernel graph cut method provides good performance, its performance is highly dependent on the data mapping to the transformation space and image features. The entropy-based kernel graph cut method is suitable for segmentation of textured images. Nonetheless, its segmentation quality remains significantly contingent on the accuracy and richness of feature space representation and kernel centers. This paper introduces an entropy-based kernel graph cut method, which leverages the discriminative feature space extracted from SqueezeNet, a deep neural network. The fusion of SqueezeNet’s features enriches the segmentation process by capturing high-level semantic information. Moreover, the extraction of kernel centers is refined through a weighted k-means approach, contributing further to the segmentation’s precision and effectiveness. The proposed method, while exploiting the benefits of suitable computational load of graph cut methods, will be a suitable alternative for segmenting textured images. Laboratory results have been taken on a set of well-known datasets that include textured shapes in order to evaluate the efficiency of the algorithm compared to other well-known methods in the field of kernel graph cut. [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]