Stereo Matching Distortion Correction for Dual Camera Systems

The patent applies preliminary action by calculating distortion parameters and performing distortion correction on images before stereo matching is executed. The distortion correction unit pre-processes the images from cameras with different optical systems or line scan speeds, ensuring that the images are geometrically aligned prior to feature point matching. This preliminary correction prevents accuracy degradation in subsequent distance estimation.

The patent utilizes parameter changes by dynamically calculating distortion parameters based on camera-specific characteristics such as angle of view and line scan speed. These parameters are used to transform the images into a common reference frame, adjusting the geometric parameters of the images to compensate for optical differences between cameras, thereby maintaining high stereo matching accuracy.

The patent replaces the mechanical sensor-based distortion detection system with a computational image processing approach. Instead of using acceleration sensors to physically detect camera movement and infer distortion, the system uses image processing algorithms to directly calculate distortion parameters from the image data itself, eliminating the need for additional mechanical sensing hardware and reducing system cost.

The patent applies self-service by enabling the image processing system to automatically detect and correct its own distortion without external sensing devices. The distortion correction unit uses the image data from the cameras themselves to calculate distortion parameters and perform correction, making the system self-diagnosing and self-correcting without requiring separate detection hardware.