Computer-aided medical image segmentation helps to assist physicians in locating lesion area for the subsequent diagnosis and treatment. Due to the irregular shape of the target and the uneven sample size between the target and the background area, automatic segmentation of medical images is a challenging task. Many CNN-Based, Transformer-Based models deepen the number of network layers or introduce complex modules in order to improve the segmentation accuracy. Limited by the computational resources, these types of large models are not suitable for the actual clinical environment. Inspired by the rapidity, accuracy, and low consumption characteristics of bio-visual processing, the Ultra-Lightweight Network Inspired by Bio-Visual Interaction (BVI-Net) is constructed in this paper. The Global Pathway is constructed by simulating the dorsal stream, in order to extract global features rapidly, and the Local Pathway is constructed by simulating the ventral stream, in order to process local features finely. At the same time, the skip connection module integrating Graph Convolutional Network (GCN) attention mechanism is constructed to simulate the synchronous integration ability of the visual pathway for multi-level features. The International Skin Imaging Collaboration (ISIC) dataset, the Liver Tumor Segmentation (LiTS) dataset, and the Brain Tumor Segmentation Challenge (BraTS) dataset are used for experiments. The BVI-Net proposed in this paper requires only 0.026M parameters to achieve the excellent performance in three representative medical image segmentation datasets, which has certain advantages over state-of-the-art (SOTA) methods. The biological vision mechanism and the artificial intelligence algorithm are integrated in this paper, which provides new ideas for the construction of biological vision-guided deep learning models and promotes the development of biomimetic computational vision.
The dehazing of images shot in fog is a hot spot in the study of computer vision. Unlike dehazing methods, which use an atmospheric scattering model, the method proposed here is based on fusion coding of contours and colors. It simulates the characteristics of visual perception in blurred scenes and balances the amount of color information in foggy images by actively fusing the contour features, thus preventing issues that often arise in dehazing, such as distortion or halo effects. First, the method constructs a contour feature extractor and extracts the contour features of the image, enhancing their weight in feature coding, and then it constructs a low-level feature-coding region to extract colors while adding the contours to fusion code the contours and colors. Then, the an advanced semantic coding region is constructed with dilated convolution residual blocks to deeply analyze the semantic information from the back propagation. Finally, after the fusion of the outputs of the low-level feature coding and the middle and final outputs of the advanced semantic coding, the method decodes the contours and colors in several layers of a convolutional neural network. This paper discriminates the results of dehazing and the sample labels using the discriminator net, composed of several convolution layers, and then intensifies and generates the network’s dehazing ability while improving the discrimination ability of the discriminator net. Synthetic and natural foggy images are chosen as experimental objects, and the results of the novel methods are compared with those of currently available methods. The results show that, unlike other methods, this method produces good results, is robust, addresses issues like color distortion and halos, and dehazes images with natural saturability and sharpness, providing new potential for study of image dehazing.
1 June 2024 in Biocybernetics and Biomedical Engineering
Vehicle detection can provide volumes of useful data for city planning and transport management. It has always been a challenging task because of various complicated backgrounds and the relatively small sizes of targets, especially in high resolution satellite images. A novel model called joint-layer deep convolutional neural networks (JLDCNNs), which joins features in the higher layers and the lower layers of deep convolutional neural networks (DCNNs), is proposed in this paper. JLDCNNs aim to cover different scales and detect vehicles from complex satellite images rapidly by overcoming the insufficient feature extraction of traditional DCNNs. The model is evaluated and compared with traditional DCNNs and other methods on our challenging dataset which includes 20 high resolution satellite images (including over 2400 vehicles) collected from Google Earth. JLDCNNs significantly improve the precision rate by 16% and the recall rate by 6% compared with traditional DCNNs, let alone outperform other traditional methods.
23 October 2025 in Biomedical Signal Processing and Control