2results about How to "High code rate" patented technology

This application discloses an HDR video coding method, apparatus, device, and medium based on texture-luminance perceptual masking, relating to the field of video coding. The method includes: converting the linear luminance average value of the area to be encoded to the target luminance domain to obtain a first logarithmic value; determining a luminance masking component based on the first logarithmic value; performing a frequency domain transformation on the prediction residual of the area to be encoded to obtain frequency domain coefficients, accumulating the energy of the frequency domain coefficients to obtain an energy value, and converting the energy value into a texture masking component; mapping the luminance masking component and the texture masking component to the same numerical interval, and fusing the two components after unifying the interval to obtain a perceptual factor; adjusting the original Lagrange multipliers based on the perceptual factor, constructing a rate-distortion cost function using the adjusted Lagrange multipliers, and performing video coding processing on the area to be encoded according to the rate-distortion cost function to obtain the encoded HDR video. This application significantly improves coding quality and efficiency.

This invention discloses a method and system for video evidence collection based on the shared usage time of instruments. By acquiring instrument identifiers and binding them to the evidence collection device, the Region of Interest (ROI) of the screen display area, status indication area, and operation area is calibrated, and a unified evidence collection timeline is established based on network time synchronization and a local secure clock. Video streams covering the ROI and auxiliary evidence streams are acquired synchronously. Multi-source data is input into a recognition model containing a multi-branch visual encoder for the ROI, an auxiliary evidence encoder, a cross-modal alignment gating module, a hierarchical temporal transformer, and an event query header. The model outputs a state probability sequence and regresses the start and end event boundaries and confidence intervals. The net usage time is calculated under state transition constraints and verified for consistency with auxiliary evidence. When the difference exceeds a threshold, frames are added to the event window for re-inference to update boundaries or mark disputes. Key segments and keyframes are extracted to generate evidence packages, a frame-level hash chain is constructed, and signatures are completed within a TEE. Combined with a timestamp service encapsulation, verifiable and tamper-proof evidence results are achieved.