Image target identification method used in underwater complex environment

Abstract

The invention discloses an image target recognition method used in an underwater complex environment. The method comprises the following steps: 1, preprocessing a picture by adopting gamma transformation; 2, extracting picture direction gradient histogram features; 3, performing the feature selection according to the picture fuzziness; 4, training and testing by using a support vector machine; 5,carrying out image identification by utilizing the support vector machine trained in the step 4. The method can overcome the difficulties of few underwater image libraries, complex picture backgroundsand few ideal underwater pictures at present, reduces the requirements of underwater image target recognition by using a support vector machine for images, can also ensure rapidity and accuracy, reduces the target feature dimension, avoids the overfitting caused by the fact that the number of samples is far smaller than the number of feature dimensions when machine learning is used for target recognition, and improves the speed and accuracy of underwater target recognition.

Description

technical field [0001] The invention relates to an image target recognition method used in underwater complex environments, belonging to the field of underwater image target recognition. Background technique [0002] Underwater image target recognition is an important branch in the field of image recognition and has broad application prospects. Using traditional machine learning methods for underwater image target recognition does not require a large amount of learning data, which can well solve the problem of insufficient sample data available for underwater images , but at the same time it also brings new troubles. In order to accurately propose target features from a small amount of learning data, it is required to have as few disturbing features as possible in the picture, such as background and lighting, and to have high quality requirements for the picture, otherwise it will affect The accuracy of image target recognition. In order to enable image target recognition t...

AI Technical Summary

Problems solved by technology

[0002] Underwater image target recognition is an important branch in the field of image recognition and has broad application prospects. Using traditional machine learning methods for underwater image target recognition does not require a large amount of learning data, which can well solve the problem of insufficient sample data available for underwater images , but at the same time it also brings new troubles. In order to accurately propose target features from a small amount of learning data, it is required to have as few disturbing features as possible in the picture, such as background and lighting, and to have high quality requirements for the picture, otherwise it will affect The accuracy of image target recognition

Images