Feature extraction and fusion recognition of dual-source images based on convolution neural network

Abstract

The invention discloses a dual-source image feature extraction and fusion identification method based on a convolution neural network, which comprises the following steps: utilizing the characteristics of the convolution neural network with migration learning, training the convolution neural network model parameters through a large number of visible light databases; The trained model is used to automatically extract the hidden features of visible and thermal infrared target images, and the maximum desampling method is used to reduce the feature dimension. Combining Fisher discriminant method and principal component analysis algorithm, the dimension reduction and fusion of multi-source image features are carried out. Support vector machine classifier is used to classify and recognize the fusion features of the target image. A method for classify and identifying multi-source image target in unmanned aerial vehicle (UAV) platform features that image hidden features are extracted by convolution neural network, and Fisher discriminant method and principal component analysis algorithm are combined for dimension reduction and fusion of features, which provides a new and effective way forclassifying and identifying multi-source image targets based on feature level.

Description

technical field [0001] The invention belongs to the field of image signal processing and pattern recognition, and is a dual-source image feature extraction and fusion recognition method based on a convolutional neural network. Background technique [0002] In the past two decades, feature-based target recognition and classification technology has become a hot spot in the research of image signal processing and pattern recognition, and has been widely used in military and civilian fields. For example, sea ship detection, sea rescue, ground military target strike, suspect tracking, etc. At present, the single-sensor feature-level target recognition and classification technology is relatively mature. Due to the limitation of the sensor itself, its working environment and applicable objects are relatively single, which cannot meet the application needs in complex environments. [0003] Visible light sensors have high imaging resolution, rich target texture details, and clear ed...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problem that the source target recognition of the traditional UAV single sensor is greatly affected by environmental factors, in order to improve the recognition efficiency and expand the scope of applicable scenarios, this invention proposes a dual-source image feature extraction and fusion recognition method based on convolutional neural network. Solve the problem that the single-source sensor used in the UAV platform for target recognition is susceptible to interference from lighting conditions, is sensitive to target changes, and has a low classification and recognition rate. provides a new way

Images