Real and Fake Label Image Classification Algorithm Based on HOG and SVM

As the market for second-hand luxury goods continues to expand, authenticity must be verified when trading. In order to solve the problems of low efficiency and high cost of manually identifying luxury goods, this paper proposes a luxury identification method based on HOG feature extraction and SVM. This method uses a black label image of Gucci bags with anti-counterfeiting as an example. First, Collect true and fake sample photos, group and label the photos, make positive and negative samples, and perform normalization processing, and then extract the true and fake label HOG features, and then train the SVM as a true and fake label image classifier, using 200 test experiments were performed on the test samples, and the experimental results show that the accuracy rate of true and fake label image classification of this method reaches 90.25%, which can well identify the true and false labels under different lighting conditions.