Multi-label-based image recognition method

Abstract

The invention discloses a multi-label-based image recognition method which comprises the steps of: preprocessing of an image sample, image segmentation, characteristic extraction, characteristic value discretization, mining of a frequent item set L, construction of a multi-label association classification rule (MLACR) and image reorganization. In the process of mining the frequent item set L, a novel candidate item set pruning method is adopted, through two times of pruning operations, the scale of a candidate item set is remarkably reduced, and the execution efficiency of an algorithm is further increased, and during the construction of the MLACR, a reduction method is used, so that excessive rules do not occur in the MLACR. The method can be use for recognizing the single image including a plurality of labels once, and can also be used for constructing the candidate item set rapidly, thus the function of accurately and effectively recognizing the multi-label image is realized.

Description

technical field [0001] The invention belongs to the application field of image computer analysis technology, and in particular relates to a recognition method with multi-label images. Background technique [0002] Image recognition is an important research branch in data mining technology, which aims to construct a classification function or classifier by training image sample data sets, and use the classification function or classifier to identify the label or label set of the image to be tested. In the traditional so-called multi-class single-label image recognition problem, each image data contains only one corresponding label. However, in practical applications, due to the complexity of the objective things themselves, an image may contain multiple different labels at the same time. " and other topics; in medical image recognition, a medical image can contain information related to diseases such as "diabetes" and "prostate". Unlike single-label classification problems,...

AI Technical Summary

Problems solved by technology

These methods have some shortcomings that cannot be overcome by the algorithm itself, such as: the decision tree obtained by the improved C4.5 algorithm using the divide and conquer strategy is not necessarily optimal, and the structure adjustment and performance improvement of the decision tree are also difficult; The problem with the algorithm Bp-MLL is that this method cannot observe the learning process in the middle, and the final output results are difficult to explain, which affects the credibility and acceptability of the results. At the same time, this method requires a long learning time; ML -KNN is also highlighted in the classification of high-dimensional data.

Images