A Harris corner detection method for asynchronous time domain vision sensor is presented

Abstract

The invention discloses a corner detection method based on a novel photoelectric imaging device-an asynchronous time domain vision sensor, and provides a concrete implementation step. This new imagingdevice is only sensitive to the light intensity change in the shooting scene, so its output visual information has the characteristics of real-time, very low redundancy and small amount of data. Compared with the traditional frame sampling imaging, the computation and storage requirements are greatly reduced, so it is more suitable for high-speed vision applications. The core idea of the invention is firstly to construct a continuously changing velocity surface by using the asynchronous event stream output from the visual sensor, and then to realize the event-driven corner detection on the velocity surface by using the Harris corner detection principle.

Description

technical field [0001] The invention belongs to the technical field of machine vision, and in particular relates to a Harris corner detection method applied to an asynchronous time domain vision sensor. Background technique [0002] Among all human information sources, image information accounts for more than 80%. At present, silicon-based semiconductor image sensors (including CCD and CMOS image sensors) have completely replaced silver iodide film and become the most important photoelectric imaging device. Based on the digital images it generates, computer vision and machine vision have been used more and more widely, and have become an important part of artificial intelligence. [0003] According to the imaging principle, the currently used images are generated based on the "frame sampling" method: [0004] 1. After all pixels are reset at the same time, they start to receive light (collect photoelectric charges), and stop receiving light after reaching the set exposure ...

AI Technical Summary

Problems solved by technology

These redundant information also put huge pressure on the processing and storage of the system;

[0010] (2), high latency

[0017] The traditional method of using the Harris principle for event domain corner detection is to calculate the time gradient of each arrival event on the time surface (the space surface formed by the event arrival time), and then calculate the corner point. The problem with this method is that when there is an object When occluding each other, false corners caused by boundary intersections cannot be distinguished, which is not conducive to accurate target tracking

Images