Retinal ganglion cells (RGCs) are highly sensitive to changes in contrast, which is crucial for the detection of edges in a visual scene. However, in the natural environment, edges do not just vary in contrast, but edges also vary in the degree of blur, which can be caused by distance from the plane of fixation, motion, and shadows. Hence, blur is as much a characteristic of an edge as luminance contrast, yet its effects on the responses of RGCs are largely unexplored.We examined the responses of rabbit RGCs to sharp edges varying by contrast and also to high-contrast edges varying by blur. The width of the blur profile ranged from 0.73 to 13.05 deg of visual angle. For most RGCs, blurring a high-contrast edge produced the same pattern of reduction of response strength and increase in latency as decreasing the contrast of a sharp edge. In support of this, we found a significant correlation between the amount of blur required to reduce the response by 50% and the size of the receptive fields, suggesting that blur may operate by reducing the range of luminance values within the receptive field. These RGCs cannot individually encode for blur, and blur could only be estimated by comparing the responses of populations of neurons with different receptive field sizes. However, some RGCs showed a different pattern of changes in latency and magnitude with changes in contrast and blur; these neurons could encode blur directly.We also tested whether the response of a RGC to a blurred edge was linear, that is, whether the response of a neuron to a sharp edge was equal to the response to a blurred edge plus the response to the missing spatial components that were the difference between a sharp and blurred edge. Brisk-sustained cells were more linear; however, brisk-transient cells exhibited both linear and nonlinear behavior.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1017/S0952523810000064 | DOI Listing |
Sensors (Basel)
November 2024
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China.
The underwater imaging process is often hindered by high noise levels, blurring, and color distortion due to light scattering, absorption, and suspended particles in the water. To address the challenges of image enhancement in complex underwater environments, this paper proposes an underwater image color correction and detail enhancement model based on an improved Cycle-consistent Generative Adversarial Network (CycleGAN), named LPIPS-MAFA CycleGAN (LM-CycleGAN). The model integrates a Multi-scale Adaptive Fusion Attention (MAFA) mechanism into the generator architecture to enhance its ability to perceive image details.
View Article and Find Full Text PDFSci Rep
November 2024
School of Digital and Intelligent Industry, Inner Mongolia University of Science & Technology, Baotou, 014010, China.
Automated segmentation of liver tumors on CT scans is essential for aiding diagnosis and assessing treatment. Computer-aided diagnosis can reduce the costs and errors associated with manual processes and ensure the provision of accurate and reliable clinical assessments. However, liver tumors in CT images vary significantly in size and have fuzzy boundaries, making it difficult for existing methods to achieve accurate segmentation.
View Article and Find Full Text PDFSensors (Basel)
October 2024
Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China.
In order to solve the problem that existing visible and infrared image fusion methods rely only on the original local or global information representation, which has the problem of edge blurring and non-protrusion of salient targets, this paper proposes a layered fusion method based on channel attention mechanism and improved Generative Adversarial Network (HFCA_GAN). Firstly, the infrared image and visible image are decomposed into a base layer and fine layer, respectively, by a guiding filter. Secondly, the visible light base layer is fused with the infrared image base layer by histogram mapping enhancement to improve the contour effect.
View Article and Find Full Text PDFSensors (Basel)
October 2024
School of Electrical Engineering and Automation, Xiamen University of Technology, Xiamen 361024, China.
Small object detection, which is frequently applied in defect detection, medical imaging, and security surveillance, often suffers from low accuracy due to limited feature information and blurred details. This paper proposes a small object detection method named YOLO-DHGC, which employs a two-stream structure with dense connections. Firstly, a novel backbone network, DenseHRNet, is introduced.
View Article and Find Full Text PDFIEEE Trans Neural Netw Learn Syst
November 2024
Prohibited item detection in X-ray images is one of the most essential and highly effective methods widely employed in various security inspection scenarios. Considering the significant overlapping phenomenon in X-ray prohibited item images, we propose an anti-overlapping detection transformer (AO-DETR) based on one of the state-of-the-art (SOTA) general object detectors, DETR with improved denoising anchor boxes (DINO). Specifically, to address the feature coupling issue caused by overlapping phenomena, we introduce the category-specific one-to-one assignment (CSA) strategy to constrain category-specific object queries in predicting prohibited items of fixed categories, which can enhance their ability to extract features specific to prohibited items of a particular category from the overlapping foreground-background features.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!