CN-122024022-A - Method and computing device for generating image forgery analysis report

CN122024022ACN 122024022 ACN122024022 ACN 122024022ACN-122024022-A

Abstract

A method and computing device for generating an image forgery analysis report. The method for generating the image forgery analysis report comprises the steps of obtaining forgery analysis results of an image to be detected and a customer portrait, and generating the image forgery analysis report with the content style consistent with that of the customer portrait based on the forgery analysis results and the customer portrait.

Inventors

ZENG FANWEI
LI JIANSHU
YAO WEIBIN

Assignees

蚂蚁区块链科技(上海)有限公司

Dates

Publication Date: 20260512
Application Date: 20260120

Claims (10)

1. A method of generating an image forgery analysis report, comprising: Obtaining a fake analysis result of an image to be detected and a customer portrait; And generating an image forgery analysis report with the content style consistent with the client portrait based on the forgery analysis result and the client portrait.
2. The method of claim 1, wherein generating an image forgery analysis report with a content style consistent with the customer representation based on the forgery analysis result, the customer representation, comprises: Disassembling the counterfeit analysis result into a plurality of evidence atoms; based on the customer representation, the evidence atom is rewritten so that the content style of the rewritten evidence atom is consistent with the customer representation.
3. The method of claim 2, wherein the method further comprises obtaining layout constraint information, Wherein based on the forgery analysis result and the customer representation, generating an image forgery analysis report with a content style consistent with the customer representation, further comprising: planning typesetting areas of the evidence atoms on the premise of meeting the format constraint information to obtain layout information; and rendering the rewritten evidence atoms according to the layout information to obtain the image forgery analysis report.
4. A method according to claim 3, wherein planning the typesetting area of the evidence atom on the premise of meeting the layout constraint information comprises planning the typesetting area of the evidence atom according to the priority of the evidence atom on the premise of meeting the layout constraint information, wherein the size and/or typesetting order of the typesetting area of the evidence atom is positively correlated with the priority of the evidence atom.
5. A method according to claim 3, wherein the method further comprises: under the condition that the evidence atom exceeds the planned typesetting area in the rendering process, compressing the text content in the evidence atom to enable the compressed evidence atom to adapt to the typesetting area, or Under the condition that the evidence atom exceeds the planned typesetting area in the rendering process, the size of the typesetting area is enlarged, and the sizes of other typesetting areas positioned behind the typesetting area are reduced.
6. The method according to claim 1, wherein the falsification analysis result includes positional information of an abnormal region in the image to be detected, the method further comprising: Acquiring the image to be detected; Performing highlighting processing on an image part corresponding to the position information in the image to be detected; And adding the processed image to be detected into the image forgery analysis report.
7. The method of claim 1, wherein generating an image forgery analysis report with a content style consistent with the customer representation based on the forgery analysis result, the customer representation, comprises: constructing a prompt word, wherein the prompt word comprises the counterfeiting analysis result and the client portrait, and the prompt word indicates a multi-mode large language model to generate an image counterfeiting analysis report with the content style consistent with the client portrait based on the counterfeiting analysis result and the client portrait; and inputting the prompt word into the multi-modal large language model.
8. The method of claim 7, wherein, The falsification analysis result includes positional information of an abnormal region in the image to be detected, The prompting word also comprises the image to be detected, and indicates the multi-mode large language model to highlight the image part corresponding to the position information in the image to be detected, and adds the processed image to be detected into the image falsification analysis report.
9. The method of claim 7, wherein, The prompt word comprises a first prompt message, a second prompt message and a third prompt message, The first prompt information indicates the multi-modal large language model to generate a plurality of evidence atoms with content styles consistent with the customer portrait based on the falsification analysis result, The second prompt information indicates that the multi-modal large language model plans the typesetting region of the evidence atom according to the priority of the evidence atom on the premise of meeting the layout constraint information to obtain layout information, And the third prompt information indicates the multi-mode large language model to render the evidence atoms according to the layout information to obtain the image forgery analysis report.
10. A computing device comprising a memory having executable code stored therein and a processor, which when executing the executable code, implements the method of any of claims 1-9.

Description

Method and computing device for generating image forgery analysis report Technical Field The embodiment of the specification belongs to the field of image evidence obtaining, and particularly relates to a method and computing equipment for generating an image forgery analysis report. Background The core goal of the image evidence collection (Image Forensics) is to carry out scientific verification and analysis on the authenticity, the integrity, the source and the historical operation of the image so as to judge whether the image is tampered and forged or trace back the generated and transmitted related information, thereby providing objective evidence for scenes such as judicial evidence collection, news check, copyright protection, public security and the like. The image forensic results are ultimately presented to the customer. However, the existing image evidence obtaining technology focuses on how to more comprehensively and accurately detect the authenticity of an image, but ignores the audience of the image evidence obtaining result, so that the image evidence obtaining result lacks audience perception. Therefore, it is desirable to have a solution that enhances the audience perception of the image forensic results that are ultimately presented to the customer. Disclosure of Invention A first aspect of the specification provides a method for generating an image forgery analysis report, which comprises the steps of obtaining forgery analysis results and customer portraits of an image to be detected, and generating the image forgery analysis report with the content style consistent with the customer portraits based on the forgery analysis results and the customer portraits. A second aspect of the present specification provides a computing device comprising a memory having executable code stored therein and a processor which when executing the executable code implements the method of the first aspect. According to the method provided by the embodiment of the specification, the image forgery analysis report with the content style consistent with the client portrait is generated based on the forgery analysis result and the client portrait, so that the image forgery analysis report with different content styles can be generated for different client groups based on the same forgery analysis result, and the reporting style of thousands of people and thousands of sides is realized. Drawings In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings that are needed in the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present disclosure, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. FIG. 1 is a schematic diagram of a method of generating an image forgery analysis report in one embodiment of the present disclosure; FIG. 2 is a flow chart of a method of generating an image forgery analysis report in an embodiment of the present disclosure; FIG. 3 is a timing diagram of a method for generating image forgery analysis reports based on a distributed system architecture implementation in an embodiment of the present description. Detailed Description In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure. With the development of image forensic technology, high quality image forgery analysis data can be obtained, however, the process of converting image forgery analysis data into a final delivery (i.e., report) still has the following bottleneck. 1. Lack of audience perception. The confidence coefficient parameters need to be displayed for technicians, evidence chain logic needs to be displayed for judges, and the users at the C end need to be easily understood. The existing scheme cannot automatically perform "speaking style migration". 2. The template is stiff and cannot adapt to the evidence density. Some pictures were tampered with only 1 and some were 20. The fixed stencil may cause the former to leave too much white and the latter to type out or the word too small. 3. And (5) typesetting and cutting. Text generation is separate from picture cropping. The situation that the characters are written for a long time, but the matching diagram is very small and cannot be seen clearly, or the mat