Research on depth image tampering detection based on color space fusion

With the rapid development of image processing technology, image tampering has become increasingly common, posing serious challenges to information security. This paper proposes a depth image tampering detection method based on multicolor space fusion, which enhances the model's ability to perceive tampering features by combining information from RGB and YCbCr color spaces. ConvNeXt is used as the backbone network for feature extraction, and effective fusion of color space features is achieved through feature splicing. Experimental results on the comprehensive data set show that the proposed method reaches more than 99% in accuracy and AUC indicators, proving the effectiveness and reliability of the method in image tampering detection tasks. Future research will focus on further improving the model's detection capabilities in complex scenes.