Local Occlusion Target Contour Repair Algorithm Based On The Gestalt Rules

The target contour can be obtained by connecting the detected boundary points.Target contour is widely used in the field of target recognition due to its advantages of simple calculation and good stability.However,in practical application,limited by camera equipment,it is difficult to guarantee the quality of acquired images.For example,there are often mutual occlusion or local missing of the target area in the image,which brings a lot of inconvenience to the recognition process.Therefore,it is of great significance to study the contour restoration of partially occluded objects in images.The results can be used in recognition of local occlusion objects,image restoration and retrieval,image blanking,and ancient cultural relic restoration.At the same time,it is also an important algorithm in graphics rendering.In order to reduce the complexity of partial shade the target contour restoration algorithm steps,and improve the accuracy of local keep out target outline repair in accordance with human visual aesthetic and the maximum extent,in this article,through further study of gestalt completion rules in partial shade the role of target contour detection and repair,based on the rules of gestalt partial shade the target contour restoration algorithm research,the main research contents and results are as follows:1.In this paper,a local occlusion target contour detection algorithm based on significance constraint and gestalt rule is proposed.The algorithm firstly improves the traditional Canny algorithm,that is,the image denoising is carried out by using the improved median filter instead of gaussian filter,and the high threshold is calculated by the improved Otsu algorithm.Then,in the improved Canny operator framework,the significance information of RC algorithm was added as a constraint to reduce the edge on the background,and then the significant edge was extracted.Then,on the basis of the Ratio-contour algorithm,a new objective function was proposed in combination with the closure in the gestalt rule,so that the algorithm in this paper could converge to a region with high significance and detect a more complete closed contour.Finally,on the basis of the extracted contour,type node detection is carried out to find the occlusion points,and the occluded target contour is segmented and extracted.The experimental results show that the algorithm overcomes the defect that the traditional Canny algorithm cannot get the high threshold value adaptively and needs to set the parameters manually,and also effectively improves the recall and precision of contourdetection.2.By analyzing the existing contour restoration methods,it is found that there are few researches on contour restoration of partially occluded objects,and none of the existing algorithms can well integrate human visual cognitive characteristics.Therefore,this paper proposes a local occlusion target contour restoration algorithm based on gestalt completion rules.The algorithm first determines the connectability of type a nodes according to their nearness and orientation similarity,and then generates the connectability graph.Then,close neighbor principle,continuity principle and similarity principle of gestalt rules are applied to group type nodes,and then connect type nodes with connectable relation but without actual connection.Finally,the polynomial fitting curve based on the least damped square method is used to connect the type nodes,and then the local occlusion target contour is repaired.Experimental results show that the algorithm is effective and accords with human visual perception.