Identification and space positioning device and method for fished targets based on omni directional vision

Abstract

An identification and space positioning device for fished targets based on an omni directional vision comprises an ODVS (omni directional vision sensor), and a microprocessor for analyzing and processing the identification and space positioning of fished targets. The ODVS comprises two fixed single viewpoint dead-angle-free omni directional vision sensors with the same parameter. The two omni directional vision sensors are fixedly connected in a perpendicularly back-to-back manner according to an omni polar plane requirement. The invention further discloses an identification and space positioning method for fished targets based on the omni directional vision. The invention adopts the ODVS to solve the problem about image acquisition for identification and space positioning of deep sea fished targets, can provide the omni directional vision for fishing robot navigation at the same time, and has the advantages of low calculated amount for stereo matching of feature points, high spatial orientation and the like.

Description

Technical field [0001] The invention belongs to the application of panoramic stereo vision technology, fishing robot technology, digital image processing technology and pattern recognition technology in the automatic fishing of seabed organisms, and is especially suitable for automatic fishing operations of fishing robots. Background technique [0002] Deep-sea creature fishing is the fishing of deep-sea creatures, and the creatures are captured from the deep sea for scientific research or other commercial purposes. At present, a trawl catching technique is usually used, which is a drag fishing method that uses ships to navigate. Bottom trawl is a technology used to capture small creatures such as deep sea creatures. This bottom trawl technology has caused catastrophic damage to the ecosystem. Corals, sponges, fish and other animals will all be killed. At the same time, underwater ecosystems such as seamounts, the habitat of many marine organisms, have also been severely damage...

AI Technical Summary

Problems solved by technology

[0010] However, the current stereo vision system still has many problems in the practical application of seabed biological fishing. The main problem is that the stereoscopic imaging range is limited. Since the fishing robot walks in the seabed fishing crop environment, the growth area of ​​the fishing object is open. Therefore, it is necessary to rotate the stereo vision system to obtain the video information of the surrounding fishing objects; secondly, there are still gaps in the practical application of stereo imaging camera calibration, feature extraction and stereo image matching. some intractable problems

In addition, because the fixed focal length can only clearly capture images within a certain depth of field, it limits its application in some aspects.

In addition, there are still the following deficiencies: 1) Due to the imperfect calibration technology, the internal and external parameters of the camera in the stereo vision measurement system are inevitable during the movement of the fishing robot, thus limiting many applications; 2) In various shapes In the process of from X, multiple coordinate conversions are involved, resulting in too much calculation and difficult real-time processing, which limits the real-time identification of fishing objects and other applications; The error affects the matching accuracy

Stereo matching is a typical visual "morbid" calculation problem. It is still difficult to perform unambiguous and high-accuracy stereo matching on images with current technology.

Images