Research on vision and error analysis of the hotte

Research on vision and error analysis of micromanipulation robot (1)

with the rapid development of nanotechnology, the research object continues to develop towards micro refinement. The processing, adjustment and cost inspection of micro parts, and the assembly of micro electro mechanical systems (MEMS) all need the participation of Micromanipulation robots. The market demand for the heat dissipation shell of LED lamps is also climbing rapidly. In the fields of adaptive optics, optical fiber docking, medicine, biology, especially animal and plant genetic engineering, agricultural product improvement and breeding, it is necessary to complete the fine operations such as injection cell fusion and micro surgery, which are inseparable from the high-precision micro operation robot system. In a word, micromanipulation robot is an indispensable tool for people to explore the micro world

micromanipulation robot system is generally composed of the following aspects: high frequency doubling, high-resolution micro vision system; High precision and human range operation with more than two degrees of freedom - a platform and auxiliary equipment; A multi degree of freedom manipulator capable of changing the pose of the operating object; Micromanipulators suitable for the operation of small objects, such as grippers for micro parts assembly, syringes and scalpels for micro surgery, etc

1 image processing

1.1 basic image processing

gray scale processing: for the convenience and real-time requirements of processing, first turn the color image into a gray-scale image, and then carry out

processing and analysis, such as generating gray-scale histogram, gray-scale repair, extracting image features, image impact testing machine, sharpening of testing instruments used to measure the impact resistance of metal materials under dynamic load, etc. The grayscale of digital image must affect the normal work, which is the basis of image recognition and processing (as shown in Figure L)

binarization processing: the gray value of the image obtained by gray processing is evenly divided into 256 copies. For the image with single cell background information, the frequency of a certain gray level will be low. If these gray levels are incorporated into the adjacent gray levels, their contrast can be effectively enhanced. Generally, the threshold value is determined by calculating the gray scale flail, and the image is divided into two parts according to the person of the threshold value, that is, the binary processing method

image nail slip: the purpose of image nail slip is to eliminate image noise. Generally speaking, the energy of the image is mainly concentrated in its low-frequency part, the frequency band of the noise is mainly in the high-frequency part, and the edge information to be extracted in the system is also mainly concentrated in its high-frequency part. In order to remove noise, it is necessary to smooth the image, and low-pass filtering can be used to remove high-frequency interference. The common method of image nail slip is. Nail slip filtering (as shown in Figure 2) or median filtering (as shown in Figure 3)

image sharpening: image nail slip processing is to blur the boundary and contour in the image. In order to reduce this adverse effect, we need to use the image sharpening technology that can compensate the edge, highlight the edge information and make the image edge clear. Jinan assaying machine sells medium clear images at a low price. The essence of image sharpening is to enhance the high-frequency components of the original image

1.2 image edge processing

contour extraction: the so-called contour extraction is that if a point in the image is black and the eight adjacent points around it are black, it means that the point is an internal point and can be deleted. In other cases, it means that this point is a boundary point and can be retained. The result of contour extraction is hollowing out the internal points of the figure

image edge connection: edge connection refers to the process of forming an ordered edge table from an ordered edge table. Edge tracking is to find edge points in order to track the boundary. Contour tracking and edge connection belong to different edge extraction algorithms. For an image, contour tracking has the effect of edge detection, but it is not very good for images with low contrast. For the edge that has been extracted, contour tracking is useless. We must use the edge connection algorithm to complete the edge extraction. The basic steps of image edge processing and the flow of image edge processing. As shown in Figure 4 and figure 5 respectively