Critical region detection based accurate complex target identification method

Abstract

The invention relates to a critical region detection based accurate complex target identification method. The critical region detection based accurate complex target identification method comprises the following steps: using a cross training method to fuse and train the whole neural network, using the convolutional neural network to extract the target characteristic, using a detection sub-networkto detect a critical region of a complex target by taking an anchor block as a reference, using a regional standard pond to convert a critical region pond into a characteristic pattern at fixed size,classifying the critical region by using a classification sub-network, and fusing the classification results of various critical regions to accurately identify the target. The whole network comprisesthe critical region detection sub-network and the critical region classification sub-network, the detection sub-network detects the critical region having a distinction degree of the complex target, the classification sub-network classifies the critical region, the classification results of the various regions are fused to identify the whole target. The two sub-networks share the characteristic extracted by the VGG convolutional neural network, so that the effect of rapidly accurately identify the complex target is achieved.

Description

technical field [0001] The invention relates to image processing technology, in particular to a method for accurately recognizing complex targets based on key region detection. Background technique [0002] The classification and recognition of complex objects is an important and basic task in the field of computer vision. Most of the parts of different types of complex targets are the same or similar, but the differences are often reflected in some key local areas. Therefore, there are a lot of interference and redundant information in the images of complex targets. However, some existing classification and recognition methods for complex targets have the problem of low accuracy because they cannot remove the interference and redundant information in complex target images. In order to achieve accurate classification and recognition of complex targets, it is of great significance to study a method for precise recognition of complex targets based on key region detection. C...

AI Technical Summary

Problems solved by technology

However, some existing classification and recognition methods for complex targets have the problem of low accuracy because they cannot remove the interference and redundant information in complex target images.

Images