Methods and system for efficient processing of generic geometric correction engine

Abstract

An apparatus and method for geometrically correcting a distorted input frame and generating an undistorted output frame. The apparatus includes an external memory block that stores the input frame, a counter block to compute output coordinates of the output frame for a region based on a block size of the region, a back mapping block to generate input coordinates corresponding to each of the output coordinates, a bounding module to compute input blocks corresponding to each of the input coordinates, a buffer module to fetch data corresponding to each of the input blocks, an interpolation module to interpolate data from the buffer module and a display module that receives the interpolated data for each of the regions and stitch an output image. The method includes determining the size of the output block based on a magnification data.

Description

BACKGROUND OF THE INVENTIONTechnical Field[0001]The present invention relates to geometric correction engines and more specifically to correcting distorted images captured by lenses and geometrically correcting to generate an undistorted image.Description of Related Art[0002]Lens geometric distortion issues in a camera system are common. The distortion can be common optical distortions, such as barrel distortion, pin-cushion distortion, or fisheye distortion. A correction therefore is needed to correct these types of distortions in hardware in combination with software or firmware. Affine transformation support is added to support the image and video stabilization application, where multiple images of the same scene need to be aligned. Perspective warp is an extension to affine transform and offers additional capabilities. Perspective transformations can align two images that are captured from different camera viewpoints or locations. This can also be used to rectify the left and ri...

AI Technical Summary

Benefits of technology

The present invention is a device and technique for improving the quality of distorted video frames. It uses an external memory to store the input frame, a computer program to compute output coordinates for different parts of the frame, and a display to show the resulting video. The device also has a module to identify rare or repeated patterns in the video and adjust the output image accordingly. Overall, the invention allows for more accurate and precise video editing.

Problems solved by technology

Lens geometric distortion issues in a camera system are common.

Most digital cameras suffer from some degree of non-linear geometric distortion.

Therefore, forward mapping is not suitable for correcting the distortion.

However, if the input blocks are spread apart, multiple cycles may be required to fetch the input block due to addressing constraints and bus width.

Latency difference for fetching is different for different input blocks resulting in performance loss and poor utilization of bandwidth.

The internal memory is fixed in size and the data required to be fetched for a high magnification factor output block may require many cycles to fetch the data resulting in a high bandwidth use and loss of performance.

However, the multiple images captured from different viewpoints contain redundant information.

There is considerable loss of performance due to cycles spent in fetching and processing the redundant blocks.

Images