Image-enhanced infrared target detection method

Abstract

The invention discloses an image-enhanced infrared target detection method, which belongs to the technical field of infrared target attribute identification, and comprises the following steps of: establishing a clear infrared image data set, and setting related parameters of a target detection network, such as a learning rate, momentum and the like; training a target detection algorithm network; performing adaptive gamma transformation on the infrared image to increase the contrast of the infrared image; carrying out detail enhancement on the image by utilizing a Sobel operator edge enhancement method; and inputting the infrared image subjected to the image enhancement algorithm into a target detection algorithm to extract features, and carrying out detection classification. By introducingthe image enhancement algorithm, the problem that the convolutional neural network target detection algorithm is not obvious in infrared image extraction feature is effectively avoided, and the problems of false identification and missing identification generated by detection of an infrared image target and an environment thermal radiation intensity difference or a far and small target are solved.

Description

technical field [0001] The invention relates to the technical field of infrared target attribute recognition, in particular to an image-enhanced infrared target detection method. Background technique [0002] The main function of infrared target detection is to locate the target that people are interested in according to the input infrared image information, and to classify the positioned target in detail, and finally give a confidence score; at present, infrared target detection technology has been widely used in intelligent traffic management field, which can significantly improve the performance of traffic supervision and vehicle control in intelligent traffic management systems; in recent years, domestic and foreign scholars have conducted a lot of research on target detection, including target detection based on pixel features, target Detection, target detection based on gray singular value method, target detection based on magnetoresistive sensor and target detection m...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the deficiencies of the above-mentioned prior art, the present invention provides an image-enhanced infrared target detection method, which introduces an image enhancement algorithm into the convolutional neural network infrared target detection algorithm, uses adaptive gamma transformation, and enhances image contrast; The edge enhancement method of the Sobel operator enhances the relevant details of the image, and makes up for the problem of misidentification and missed identification when the convolutional neural network infrared target detection algorithm detects a target with a small difference in thermal radiation intensity between the target and the environment or a far small target. Improved Confidence and Accuracy of Infrared Object Detection with Convolutional Neural Networks

Images