Texture image quality estimation method based on Haar features and AdaBoost algorithm

Abstract

The invention discloses a texture image quality estimation method based on Haar features and an AdaBoost algorithm. The texture image quality estimation method comprises the following steps: step 1, sample collection: collecting a positive sample and a negative sample; step 2, sample labeling: respectively labeling the positive sample and the negative sample by using different numbers; step 3, feature selection: Haar features include multiple modes, and corresponding feature modes are selected according to different features of textures; step 4, AdaBoost cascade classifier building and parameter setting, wherein an AdaBoost algorithm is a method for combining a plurality of weak classifiers into a strong classifier, and the recognition rate and the false recognition rate of each layer of weak classifier and the number of the weak classifiers need to be specified; step 5, performing sample training to obtain positive and negative binary classifiers; and step 6, estimating the quality ofthe texture image. According to the method, a texture image is segmented into overlapped image sub-blocks, the image sub-blocks are classified by using a machine learning method, and finally, qualityestimation is obtained by combining classification results of the image sub-blocks.

Description

technical field [0001] The invention relates to the field of texture image quality estimation by using machine learning and image recognition, in particular to a texture image quality estimation method based on Haar feature and AdaBoost algorithm. Background technique [0002] The texture of an image is a representation of changes in the color or grayscale of the object's surface. The structure or color of each local area of ​​the object itself, changes in illumination, and the adjacency and transition between objects usually form a texture effect visually. Texture is one of the most important attributes of an image, and it is an important basis for distinguishing objects, segmenting different objects, and recognizing objects. Therefore, the lack of texture, artifacts, blur, noise and interference in the image will bring difficulties to the analysis and recognition of the image, and even affect the final analysis results and recognition accuracy, which often lead to serious ...

AI Technical Summary

Problems solved by technology

The feature summary of this type of method often relies on human understanding. The calculation process of feature parameters usually involves more prior parameter settings related to specific image characteristics, and the overall calculation degree is also complicated, which is not conducive to real-time processing.

Images