Background & Applications
Image processing, analysis and machine vision are one of the most significant and contributory branches in cognitive science as well as computer science. It has been stepping into a further degree in 1980s to 1990s , accompanied by increasingly practical applications in these field. Among the applications , remote sensing，technical diagnosis，autonomous vehicle navigation and medical imaging are the fastest development directions. Also applications like Photography and printing，Satellite image processing ，Face detection,feature detection, face identification， Microscope image processing ，Car barrier detection will fail to be performed well without image processing.
Techniques
To illustrate it in a conclusive word, we need technology like digital image processing，digital image analysis ，machine vision , * pattern recognition*. To be more specific, it is aimed to apperceive and understand the image electrically in a version of human beings. There is an usual way to process image :
- Image is captured by a sensor （For instance TV camera）.
- It will be digitized by the computer, and then be preprocessed .(noises will be curbed)
- Certain features might be enhanced to be better understand the image.(Edge extraction could be a case in this step)
- Image segmentation ( separate objects from the background and from each other)
- Recognition and understanding
Processed images are demonstrated by matrix with brightness values of the picture in a lower level. However, high-level data, which contributes to a remarkable decrease in the amount of data, only consists of partial data which extracted from highly relative objects. Also, high-level data differs from lower one in the form of data, which means symbolic form data has replaced the matrix.
Application
Machine vision with intelligent image processing functions is equivalent to people installing eyes on the machine while giving it intelligence, enabling the machine to " see the world" and " see accurately" , which can replace or even surpass human eyes for measurement and judgment, enabling the machine vision system to achieve high resolution and high speed control. Moreover, the machine vision system has no contact with the detected object and is safe and reliable.
Educational examination
Typographical or printing errors are often found in examination papers that will affect students. Using intelligent image processing technology, the machine automatically compares the printed papers with the original papers, and will automatically prompt and alarm if any discrepancies are found, which can only be checked manually before they are completely replaced.
Security monitoring
This is an area that has received much attention from machine vision at present. Machine vision breaks the limitation of traditional video monitoring system, increases the intelligence of the system, and enables intelligent video analysis to be gradually realized. Taking video surveillance in public places as an example, automatic detection of suspicious characters, face recognition and real-time tracking can be realized by using machine vision technology. When necessary, multiple cameras can also be tracked one after another, and alarms can be sent out to store on-site information.