Image Sensor Chip Frame Difference Computation

The patent extracts the frame difference computation function from the processor and relocates it to the image sensor chip. Specifically, the pixel array is divided into a first pixel array and a second pixel array, where the first pixel array captures the current frame and the second pixel array stores the previous frame. The difference computation is performed directly in the output circuit of the image sensor chip, eliminating the need for the processor to perform this computation and thereby reducing computational load while maintaining processing capability.

The patent segments the pixel array into multiple independent pixel arrays (first pixel array and second pixel array) that can operate simultaneously. Each pixel array independently captures and processes image data, allowing the system to perform frame difference computation in parallel without blocking the main processing pipeline, thus improving frame rate while maintaining processing capability.

The patent merges the frame buffer function with the pixel array structure itself. The second pixel array serves dual purposes: it acts as both an active pixel array for capturing image data and as a frame buffer for storing the previous frame. This integration eliminates the need for a separate frame buffer memory, reducing device complexity while maintaining the capability for continuous frame difference computation.

The patent implements multi-functionality by designing the second pixel array to serve multiple functions: it functions as an active imaging sensor during its active period and simultaneously serves as a frame buffer storing the previous frame during the active period of the first pixel array. This universal design eliminates the need for dedicated frame buffer hardware, reducing overall device complexity.