Multi-focal-length image acquisition device based on image fusion technology and analysis method

Abstract

The invention discloses a multi-focal-length image acquisition device based on an image fusion technology and an analysis method. The acquisition device comprises an image processing unit and a camera lens array unit; the image processing unit is electrically connected with the camera lens array unit; the camera lens array unit comprises a prime lens group and an image sensor. The prime lens group comprises at least two prime lenses; the adjacent optical imaging areas are overlapped; the image sensor is electrically connected with the prime lens; According to the invention, the long-distance road monitoring is subjected to segmented shooting, and adjacent optical imaging areas are spliced and fused during image processing, so clear display of monitoring images in a long-distance range is realized.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to a multi-focus image acquisition device and analysis method based on image fusion technology. Background technique [0002] At present, there are 20,000 expressway tunnels in our country, and the total length of the tunnels is about 20 million meters. The safety of the tunnels has always been a very important part of tunnel construction. In recent years, the state has been emphasizing the upgrading of road tunnels, mainly including filling up the shortcomings of tunnel traffic engineering and ancillary facilities, promoting the reform of the civil structure of road tunnels in service, so that road tunnels can better provide safe and convenient transportation for the public. travel service. The most important thing for upgrading the quality of highway tunnels is the informatization and intelligence of tunnel management, mainly for the rapid identification of image information in t...

AI Technical Summary

Problems solved by technology

Due to the rule that the near image is larger and the far image is smaller, and the field of view is divergent, the image collected by the camera is clearer at the near end, while the effective pixels occupied by the road in the far end image are less, which will easily lead to the loss of effective lane information and seriously affect the Intelligent management in the tunnel; on the other hand, because the effective pixels occupied by the remote vehicles in the image are too small, it is completely impossible to realize the adaptation of intelligent algorithm analysis, and there are false positives and unrecognizable problems in the analysis of people, vehicles, objects, and lanes. problems, it is difficult to achieve security management in the tunnel

Images