Timber defect sample obtaining method and apparatus, electronic device and storage medium

Abstract

The embodiment of the invention discloses a timber defect sample obtaining method and apparatus, an electronic device and a storage medium. The method includes the following steps: obtaining data of qualified timber samples; obtaining defect characteristic images on the basis of the data of the qualified timber samples, wherein the defect characteristic images are obtained from an existing defecttimber sample database; and combining the defect characteristic images and the data of the qualified timber samples together to obtain data of defect timber samples that have the same timber categorywith timber in the qualified timber samples. Through the above way, problems that a trained artificial intelligence model cannot identify timber defects and the identification precision is low due toloss of defect timber samples in a training initial stage of the artificial intelligence model are overcame, defect timber sample data obtaining ways are simplified, and costs are saved.

Description

technical field [0001] The present disclosure relates to the technical field of artificial intelligence, and in particular to a method, device, electronic equipment, and computer-readable storage medium for acquiring wood defect samples. Background technique [0002] Timber factories need to carry out quality inspections on wood, including sample quality grading and identification of defective samples. From the initial artificial vision inspection, to machine vision inspection, to today's machine learning inspection, each new stage has improved performance or efficiency compared with the previous old stage. Machine learning detection needs to train the neural network first, and the trained model is then used for detection. Neural network training requires training samples with labeled categorical data. The sample data is generally obtained by collecting timber images from timber factories, and then classifying and labeling by labelers. [0003] However, generally in wood ...

AI Technical Summary

Problems solved by technology

[0003] However, generally in wood processing plants, there are many qualified wood samples and few defective samples, and the unbalanced distribution of samples makes it difficult to obtain defective samples. In this way, the trained neural network can only identify qualified wood, but not for defective wood. Wood recognition ability is not high or the accuracy is not high

In addition, the characteristics of qualified samples of different types of wood are different, and a recognition model of another type cannot be directly transplanted into a new wood type or product

Images