Image tamper detection and tamper localization method based on discrete wavelet transformation
A discrete wavelet transform and tamper detection technology, applied in the field of image processing, can solve the problems of weak sensitivity and discrimination of tampering, difficulty in determining the location of tampered areas, etc.
Inactive Publication Date: 2011-10-05
XIAN UNIV OF TECH
View PDF5 Cites 6 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] (1) The robustness, tampering sensitivity and discrimination of the image signature are in conflict with each other, the stronger the robustness of the image signature, the weaker the tampering sensitivity and discrimination
[0008] (2) Compactness and tamper localization capabilities are contradictory
The shorter the signature, the less information it contains, making it more difficult to locate the tampered area
[0009] Therefore, how to achieve a balance and trade-off between robustness and tamper-sensitivity, that is, how to make the signature robust to regular image processing operations that preserve the content, while being highly sensitive to tampering operations that change the content; Designing an image signature algorithm that is both compact and capable of accurate tamper localization is a challenging research topic
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0065] The present invention will be described in detail below in combination with specific embodiments.
[0066] An image tampering detection and tampering positioning method based on discrete wavelet transform of the present invention comprises the following steps:
[0067] Image sender:
[0068] The first stage is to divide the image into blocks, extract image features by using 3-level DB2 wavelet transform for each block, and then use low-frequency wavelet coefficients to construct feature vectors. The specific method is:
[0069] Step 1. Feature extraction
[0070] Step 1.1. For an image I of size N×N 0 Blocking, the obtained image block is defined as B oi , i=0, K, (N 2 / P 2 )-1, the size of each image block is P×P, then N 2 / P 2 blocks, let P=8. with B oi to represent each block, where i=0, K, (N 2 / P 2 )-1. B oi (x, y) is defined as image block B oi The gray value of the pixel (x, y) in the center, where 0≤x, y≤P-1;
[0071] Step 1.2, for matrix B oi P...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an image tamper detection and tamper localization method based on discrete wavelet transformation. The method comprises the following steps of: by a picture sender, extracting features, performing KL transformation, performing quantitative compression to generate the original image signature, and sending the original image signature to an image and picture receiver; and by the picture receiver, generating a middle signature of an image to be detected, recovering the middle signature of the original image and continuously performing tamper detection and tamper localization. The method has higher robustness for the conventional image processing operation and has sensitivity to local tamper. By the method, malicious tamper features can be well captured and can be precisely localized.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an image tampering detection and tampering positioning method based on discrete wavelet transform. Background technique [0002] With the development of computer networks and information technology, and the rapid popularization of digital cameras and other equipment, people can obtain a large number of digital images more conveniently and quickly than ever before, and the exchange of image information has reached an unprecedented breadth. At the same time, powerful image processing and editing software has gradually become popular, such as Photoshop, AeDsee, iphoto, etc., enabling users to easily edit and tamper with the original image data to create fake effects. Therefore, how to effectively detect the integrity and authenticity of the image content has become an important hot issue and a difficult problem to be solved urgently faced by the legal circle and...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06T1/00
Inventor 王晓峰王尚平张亚玲刘真理
Owner XIAN UNIV OF TECH