Image resampling operation interpolation type identification method

Abstract

The invention provides an image resampling operation interpolation type identification method. The method comprises the following steps: establishing an image resampling operation expression model; performing an initial blind deconvolution operation on an image subjected to resampling to obtain an initial kernel; figuring out initial kernels with different sizes to obtain a group of convolution kernel set with multiple scales and different sizes; performing convolution on a convolution kernel in the convolution kernel set and an original image corresponding to the convolution kernel, and comparing the original image with the image subjected to resampling, wherein an optimal kernel with the minimum quality difference is the system output; displaying the optimal kernel H* as a column vector; putting the column vector in a classifier to train a model to obtain a classification model; and repeating the processes of step 1 to step 5 on a to-be-measured image, and testing the obtained column vector in the classification model obtained in step 6 to obtain a final detection result. The image resampling operation interpolation type identification method provided by the invention has the advantages of low detection feature dimension, capability of effectively shorten the training time of a classifier of a support vector machine (SVM), high compression resistance and excellent detection accuracy.

Description

technical field [0001] The invention relates to an image tampering detection method. In particular, it relates to an image resampling operation interpolation type identification method. Background technique [0002] With the advent of low-cost and high-resolution digital cameras and sophisticated editing software, digital images can be easily manipulated and changed. Fake digital images are often difficult to distinguish from real photos. Therefore, photographs can no longer serve as a record of the authenticity of events. Especially in the fields of judicial system and news media, the authenticity of image information plays an increasingly important role. In order to ensure the authenticity and integrity of digital information, digital image passive forensics technology came into being. Different from the active forensics technology represented by digital signature and digital watermark, the passive forensics technology is only based on the acquired digital image to fin...

AI Technical Summary

Problems solved by technology

[0005] Although the detection of resampling has been carried out for many years, there are still many problems, such as poor resistance to JPEG compression, difficulty in detection of aperiodic interpolation, and difficulty in detection of interpolation type identification in resampling operations, etc.

Images