Robot Vision Basics
For a vision system to be useful at providing data it must be able to perform 3 sequential steps. These steps involve capturing an image, enhancing the image and then extracting useful data. The sophistication of any vision system depends solely on the resolution of the image captured, the speed at which the above steps are carried out and the quality of the data extracted from the whole process. Today’s vision systems provide the hardware and software that carry out these tasks to a very high standard and this allows us to use these vision systems for the context in which they were designed.
The video camera and printed circuit board pair are used to convert the different light levels that make up the image into data that can be stored in the computer’s memory. This data can then be programmed to achieve the desired output. Converting this image into a data structure requires the image to be dissected into very small sections. Each section relates to a pixel and is dependent on the hardware used in the system. This is the smallest area that can be resolved by the hardware and the more pixels there are in an image, the higher the resolution of the image. In the vision software the capture command instructs the vision system to take the current vision image from the cameras light plane and store it in the computer’s memory.
The number of light levels corresponding to white and black is referred to as the grey levels of the system. The grey level relates to the brightness levels set on a particular pixel. The minimum level for any system is 32 grey levels, where the computer dictates five bits of brightness for each pixel. If colour is used, the amount of information explodes and becomes too large to store and manipulate. Grey levels address the brightness of the pixel only and colour represents the frequency of the pixel. For more information on vision systems please click on this link.
The video camera and printed circuit board pair are used to convert the different light levels that make up the image into data that can be stored in the computer’s memory. This data can then be programmed to achieve the desired output. Converting this image into a data structure requires the image to be dissected into very small sections. Each section relates to a pixel and is dependent on the hardware used in the system. This is the smallest area that can be resolved by the hardware and the more pixels there are in an image, the higher the resolution of the image. In the vision software the capture command instructs the vision system to take the current vision image from the cameras light plane and store it in the computer’s memory.
The number of light levels corresponding to white and black is referred to as the grey levels of the system. The grey level relates to the brightness levels set on a particular pixel. The minimum level for any system is 32 grey levels, where the computer dictates five bits of brightness for each pixel. If colour is used, the amount of information explodes and becomes too large to store and manipulate. Grey levels address the brightness of the pixel only and colour represents the frequency of the pixel. For more information on vision systems please click on this link.