4 results about "Image forensics" patented technology

This invention discloses a proactive forensic method for GIF face images based on spatiotemporal watermarking, belonging to the fields of image forensics and deep learning technology. This invention constructs a spatiotemporal adaptive residual encoder (STARE) and a deep integrity recovery decoder (DIRD), utilizing 3D convolution and adaptive attention mechanisms to capture the global spatiotemporal dependencies of the GIF, achieving coherent embedding of the watermark signal in the temporal dimension, eliminating visual flicker and enhancing resistance to temporal attacks. Simultaneously, this invention introduces an adversarial training mechanism, forcing the watermark information to be embedded in a deep spatiotemporal region robust to semantic reconstruction. Therefore, while ensuring high-fidelity visual quality, this significantly improves the watermark survival rate and forensic accuracy of GIFs when facing deepfake tampering.

This utility model provides a river dredging device for water conservancy supervision, belonging to the field of river dredging. It includes a tracked chassis, a dredging machine body mounted on the tracked chassis, a data acquisition unit, and an image evidence collection unit and a communication unit mounted on the dredging machine body. The data acquisition unit includes an acquisition chamber, a power supply located within the acquisition chamber, a cross-sectional data acquisition unit, a silt heavy metal information acquisition unit, a signal processing unit, and a signal storage and uploading unit, all electrically connected to the power supply. The acquisition chamber is located at the rear end of the tracked chassis. The signal acquisition ends of the cross-sectional data acquisition unit and the silt heavy metal information acquisition unit extend out of the acquisition chamber through sealed connectors. The signal receiving ends are respectively located within the acquisition chamber and are communicatively connected to the signal processing unit. The signal output end of the signal processing unit and the image evidence collection unit are communicatively connected to the signal storage and uploading unit. This solution facilitates the acquisition of complete river dredging data and is suitable for application in water conservancy supervision work.

This invention discloses a method for separating and identifying the authenticity of AI-generated content in distorted images, belonging to the fields of image forensics, AI-generated content detection, computer vision, and image credibility verification. This method addresses the problem that original metadata, content credentials, and generation parameters are easily lost or invalidated after images are transferred through platforms, screenshots, screen captures, or printed images. First, the method acquires the image to be detected and reads the image data, extracting propagation state features such as resolution ratio, compression marks, boundary regions, moiré patterns, paper texture, noise residuals, and file structure. Then, based on these propagation state features, the propagation state of the image to be detected is identified, and the corresponding detection branch is invoked to determine the effective content area, extracting content authenticity features and acquisition authenticity features respectively. Next, weights are assigned to various features according to the propagation state and dynamically fused to calculate the content authenticity risk result and acquisition authenticity judgment result. Finally, an authenticity separation result is generated, outputting a detection report containing the propagation state, acquisition authenticity, content authenticity, AI generation risk level, evidence items, and uncertainty explanation. This invention can distinguish between the authenticity of the image acquisition method and whether the content carried by the image has the risk of AI generation, even when the original image file information is invalid. It is applicable to the auxiliary identification of AI-generated content in scenarios such as screenshots, platform transfers, screen re-photographs, printed re-photographs, and mixed transmission distortion.

This invention discloses a dynamic forensic method and system for AI-generated images based on hyperbolic space. The method includes: extracting multi-scale features of the image to be forensic and embedding them into hyperbolic space to obtain hyperbolic embedding anchor points; generating a first forensic judgment result by calculating the distortion degree and penalty value between the anchor point and the real image prototype. Secondly, when the preliminary result is in an ambiguous range, the system dynamically triggers a cross-modal forensic verification link, performing a deep comparison combining physical geometry and semantic logic common sense to arrive at a final judgment conclusion. Finally, if the image is confirmed to be AI-generated, the system will accurately locate candidate generation engines through tangent space retrieval and utilize a proxy network to trace the contribution ranking of training data, ultimately outputting a tamper-proof structured forensic report chain. This invention effectively amplifies the hidden artifacts of high-fidelity AI images by utilizing the characteristics of hyperbolic space, achieving high-precision identification and dynamic balance of system computing power while providing complete engine tracing and data attribution capabilities.