A High-Precision Binocular Measurement System

The system uses three-dimensional measurement of binocular vision theory, through optical, mechanical, computer and other aspects of the technology to measure the target without contact, and up to a certain precision. In the high-voltage area, high-altitude target,and the target which is not easy to touch, there is a unique measure advantage. Binocular stereo vision is based on the principle of parallax and using imaging devices from different locations to obtain two images of the measured object and obtain three dimensional geometric of target by positional deviation between corresponding points in computer image. Binocular stereo currently used in four areas: robot navigation, micro operating system parameter detection, three-dimensional measurements and virtual reality. In addition, the system not only to measure the length of the distant object, the system can also be used to measure the width, surface area, height and tilt angle. As the tip of an optical imaging technology this system has a broad application prospects in the future.

Design of Program Framework of Binocular Vision Measurement System Based on DM642

As the rapid progress and development of DSP technology, a new way to solve the video signal processing was accepted. At present, binocular measurement instrument are mostly based on image capture card and PC on the market, expensive and bulky. For this problem, binocular vision measurement system based on DSP has been used. The design of elementary program structure about video processing which based on DSP/BIOS was finished and verified for its real-time. The results of the experiment show that the driver can be well used in the system.

The Research of Binocular Ranging System on Independent Mobile Robot

Timeliness and accuracy is a key to be resolved on robot binocular measurement. In this paper, a kind of robot vision projection has been completely established. It analyzes the principle of binocular ranging in three aspects, makes the calculation concise and easy to understand, and expands the range of effective distance. On binocular image processing, we have proposed a gray-scale computing for, firstly, generating characteristic area, then, executing template matching in the area, finally, extracting feature points and matching them in the templates. It ensures certain robustness to noise spots and tries its best to avoid mismatches. The experiments show that the robot vision system has a better accuracy and a low time complexity, and the robot can react in real time.

THEORY AND MEASUREMENT OF VISUAL MECHANISMS

Monocular threshold stimulus intensities (ΔIo, photons) were measured along the 0–180° meridian of human retinae for three observers. The test image was small (= 0.08°) and of short duration (= 0.20 second). ΔIo was found to decrease as the angular distance from the fovea was increased. Actual counts of the number of retinal elements per mm.2 along the 0–180° meridian (Østerberg) were compared with the obtained results. No direct correlation was found to exist between visual sensitivity and the number of retinal elements. Binocular threshold stimuli were also measured along the same meridian. The form of the function relating binocular visual sensitivity and retinal position was discovered to be essentially similar to that for monocular sensitivity, but is more symmetrical about the center of the fovea. The magnitude of the binocular measurement is in each case smaller than that of the monocular threshold stimulus intensity for the more sensitive eye. The ratio is statistically equal to 1.4 (a fact which suggests Piper's rule). These results are shown to be consistent with the hypothesis that the process critical for the eventuation of the threshold response is localized in the central nervous system. They are not consistent with the view that the quantitative properties of visual data are directly determined by properties of the peripheral retina.