Research of Some Image Filter Algorithms Used in Object Detection
Аннотация
Edge detection is an important step in object detection that allows accurate object localization and identification of objects in an image. In this paper, we present a comprehensive explanation of some edge detection filters, namely Sobel, Prewitt, Canny, Laplacian of Gaussian (LoG), and Scharr filters. This paper discusses the operation of the above filters, their mathematical representations, and their applications in the real world in different fields like security, medical imaging, and self-driving cars. The article contrasts the benefits and limitations of using each filter, i.e., their sensitivity to noise, computational complexity, and precision. The article ends with suggestions for future work that can be done to improve edge detection and its integration into more efficient and resilient object detection systems.