Abstract
Contour is a very important feature in biological visual cognition and has been extensively investigated as a fundamental vision problem. In connection with the limitations of conventional models in detecting image contours in complex scenes, a hierarchical image contour extraction method is proposed based on the biological vision mechanism that draws on the perceptual characteristics of the early vision for features such as edges, shapes, and colours. By simulating the information processing mechanisms of the cells’ receptive fields in the early stages of the biological visual system, we put forward a computational model that combines feedforward, lateral, and feedback neural connections to decode and obtain the image contours. Our model simulations and their results show that the established hierarchical contour detection model can adequately fit the characteristics of the biological experiment, quickly and effectively detect the salient contours in complex scenes, and better suppress the unwanted textures.
Highlights
Contour detection refers to the identification of the boundary between the target and the background in an image, focusing on the dividing lines outside the targets [1]
The non-classical receptive field (NCRF) introduces an important viewpoint concerning the local environmental background, and the environment can have specific features. It enlarges the receptive field of the nerve cells by multiple folds, providing an affective neural mechanism for the visual system to perceive image features under complex natural scenes. It is exactly the interactions based on the complex nonlinearity between classical receptive field (CRF) and NCRF that lead to the diversity of cell response characteristics, and when there is texture in the environment, the NCRF will generate an obvious suppression modulation effect on the CRF response, thereby reducing the response of unwanted textures in the cluttered background to highlight the boundaries of the contour region
To verify that the hierarchical contour detection model constructed in this paper can better simulate the feature perception and expression capacities of the cells at early visual stages and can better fit the biological characteristics, in connection with the more complex structures of a natural human image as shown in Figure 4, the CRF response characteristics of V1 complex cells for different areas of the image are analysed in specific experiments, and the results are as shown in Figure 4a–g, which correspond to the responses of the pixel points at positions a–g in the human image, and the polar coordinate values correspond to a group of 36 V1-cell model response values in different preferred directions
Summary
Contour detection refers to the identification of the boundary between the target and the background in an image, focusing on the dividing lines outside the targets [1]. By simulating the information processing mechanisms for the receptive fields of the cells in the early visual system, we propose a biologically inspired hierarchical contour detection (BIHCD) model to quickly and effectively detect the region boundaries in complex scenes. This paper is focused on studying the receptive fields and functional characteristics of the cells at early visual stages in combination with existing neuron models and the visual perception mechanisms, and on proposing a connection platform for the HVS and the computer vision system with the goal of detecting salient contours.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
