Outdoor multi-scene rapid classification and identification method based on hierarchical compression excitation network

Abstract

The invention discloses an outdoor multi-scene rapid classification and identification method based on a layered compression excitation network; the method comprises the steps: 1, obtaining the global information of surrounding scene segments through the layered compression excitation network according to the real-time data collected by a sensor device; 2, establishing a scene rapid classification and identification model based on global scene information; 3, performing hierarchical design on the classification and recognition model by constructing a difference matrix and constraining the number of non-zero lines to obtain an optimized network structure, so that the hierarchical compression excitation network adaptively processes the image features purposefully. The invention provides a rapid classification and identification method which considers multiple outdoor scenes and is simple in calculation method, and the operation loss of the classification and identification network after structure optimization is greatly reduced.

Description

technical field [0001] The invention relates to the field of image recognition methods based on neural networks, in particular to a rapid classification and recognition method for various outdoor scenes based on layered compressed excitation networks. Background technique [0002] The classification and recognition of outdoor scenes has broad application prospects in intelligent transportation systems and intelligent monitoring systems, but it is still a difficult problem in the field of computer vision, because outdoor scenes are more complex and have too many uncertain factors, such as pedestrians , Cars, animals, lighting, environment, etc. are highly random and cannot be equated to any typical scene. Coupled with the complex background, the accuracy of classification and recognition is not high. [0003] In the whole field of image processing, feature extraction technology is the most basic and important step. In recent years, most researchers' research on image proces...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for rapid classification and recognition of various outdoor scenes based on a layered compression excitation network, in order to be able to perform classification and recognition tasks on the extracted Features are optimized to obtain information element characteristics that are more in line with human understanding of the scene, reducing network runtime losses

Images