The research of garbage classification and recognition based on surf and geometric hashing algorithm

The classified collection and treatment of municipal solid waste is an important way to solve the problems of the rapid increase of urban waste production and the low efficiency of garbage treatment. In order to improve the efficiency of front-end collection in the process of garbage classification, the problem of fast and efficient matching of garbage has always been a research hot spot [1]. The development of machine vision technology provides a method to solve this problem. For the environment of garbage classification is more complex, the detection target may have the influence of occlusion, rotation and other aspects, leading to the detection target and template matching problems, so we must explore the corresponding solution for the specific target characteristics. In order to adapt to the extraction of feature points and target matching in complex environment, we select the more adaptive Surf algorithm [2] detects feature points and geometric hash algorithm to achieve target matching, and the two algorithms are optimized and improved. The experimental results show that the method can adapt to the complex environment such as rotation and occlusion in garbage classification, effectively achieve target matching, and improve the accuracy and speed of target recognition.