Method for evaluating alignment of sticked image and image processing apparatus

Abstract

A method includes aligning first image frames of a first image stream from a first image sensor and other first image frames of a second image stream from a second image sensor according to a first alignment, aligning second image frames of the first image stream and other second image frames of the second image stream according to the first alignment, acquiring a first stitched image frame by mixing the aligned first image frame and the other first image frame in a first mixing region according to a first mixing function, acquiring a second stitched image frame by mixing the aligned second image frame and the other second image frame in a second mixing region according to a second mixing function, and evaluating the first alignment based on a difference pattern, the difference pattern being calculated based on values of pixels of the first stitched frame and the second stitched frame. Pixels are included within at least one static portion of the first and second mixed regions in the respective first and second stitched image frames.

Description

technical field [0001] Embodiments herein relate to methods and image processing apparatuses for assessing alignment of stitched images. A corresponding computer program and computer program carrier are also disclosed. Background technique [0002] The use of imaging (especially video imaging) to monitor the public is common in many areas around the world. Areas that may require surveillance are, for example, banks, shops, and other areas that require security, such as schools and government facilities. Other areas that may require surveillance are process, manufacturing and logistics applications where video surveillance is primarily used to monitor the process. [0003] For some applications, a large field of view of the video is required, at least occasionally. This can be achieved in several ways. For example, a single-sensor camera can be equipped with pan-tilt-zoom technology that moves the camera and changes the field of view. This solution is usually quite expen...

AI Technical Summary

Problems solved by technology

This method of alignment requires additional complexity and processing power, and possibly additional bandwidth on the network over which this data is sent, compared to performing no alignment

[0009] Other alignment methods performed during the camera's runtime can affect the quality of the stitched image

The reduced quality of the stitched image can have a negative impact on the video processing done after stitching, and at the end can be perceived as a lower quality image by the end viewer

Images