Generally，image processing is the digital image processing .Digital image is used digital camera and scanner to obtain the two dimensional array after sampling and digitization.This element of the array is called pixel its value is an integer which called the gray level.The main contents of the image processing techniques include three parts image compression,enhancement 、restoration, matching 、describes and identification .
The common image processing are the image digitalization, image coding, image enhancement, image restoration, image segmentation and image analysis, etc.Some processing is achieved by optical methods or simulation technology , but the flexibility and convenience not as good as digital image processing . So digital image processing is the major of the image processing .
For the purpose of image analysis and understanding of segmentation, description and recognition which will be used in a variety of automated test system ,such as characters and graphics recognition, product assembly and inspection with robots, military target recognition and tracking, automatic fingerprint identification, X-ray and blood samples of automatic processing, etc.In this type of application, it’s necessary to the integrated application of pattern recognition and computer vision technology.
In a sense, image processing and analysis is one part of the machine vision.But in another sense, image processing and analysis is a broader discipline.However, these two areas are inseparable and supported each other in actually !
The main task of the machine vision can be divided into:
Firstly,the positioning, which judged the location of the object automatically, and output the location information through the communication protocol ;Secondly , the measurement, which measured the appearance of the product size automatically,such as the size measurement of outline shape, pore diameter, height, area etc.Thirdly, defect inspection ,This is the most frequent used function of machine vision system ,It can inspect some information on the surface of the product.Such as the right and wrong packing , packaging and printing right or wrong and whether it had the surface scratches or particle dust did, breakage, oil, plastic parts perforation, injection molding or not etc.Basically, it is necessary to judge by the human eyes and it also can replace the vision to get some product performance.
For defect inspection , that is the comparison for the object .It is compared the unit of the production line with the same type of KGU (known quality units),you can find manufacturing defects ,such as the lack lack of components or label.This kind of comparison is likely to be a simple pattern subtraction,and it also
also involve the geometry or vector graphics matching algorithm.If you are comparing the object or the direction are different , you should adopt the last one !The comparison of types including test objects, matching color and printing quality.All of these are based on machine vision inspection technology.
