More and more Frame Grabber or Add On Card and supporting software solutions are emerging in the market, we would like to share some information for your reference.


Comparison of hardware functions

The current image capture card on the market can be divided into three categories: one, vision capture cards. Its main function is to convert the analog image signal output from the camera to Engineering Digital Network, convert it into a digital signal, and finally transfer it to the memory in the computer. Second, a capture card with a display function. This kind of board card, on the basis of the previous acquisition card, has added another image display function. That is, the image can be directly displayed on any display. Third, a board with a processor. The board itself is designed with a processor, and the program for image processing can be run directly on the board without having to be in the computer.
Because the above three kinds of boards are very different in hardware, it is difficult to make a horizontal comparison. Therefore, we only analyze and compare some common functions of the board.

Camera support:
With the emergence of various new cameras on the market, the problem of matching the board to the camera becomes even more important.
     1. Standard. CCD cameras come in various formats, such as CCIR, RS170, etc. Determine whether the board supports all standard signals.
     2. Asynchronous signals. The asynchronous signal can save time for taking pictures. But before choosing a camera with an asynchronous signal function, first make sure that the board also has this function. This function is essential for systems that require extremely high operating speeds. Taking RS170 as an example, if there is no such function, the system waits for a photo to take 0 to 40 milliseconds.
     3. Progressive scan. More and more systems will use progressive scanning cameras, so we must first know whether the board supports this camera.
     4. Multi-resolution support. Not every board can support cameras of all resolutions.
     5. Multiple channels. Most boards can control more than two cameras at the same time. As for other uncommon functions, such as clock function, etc.


Comparison of software solutions


With different companies' vision software packages, how to make the right choice? It can be considered from the following aspects:
· Development hardware environment: Whether this software package can run normally under the computer hardware system selected by your company. For example, whether it supports PCI interface, display card, etc.
· Development operating system: whether this software package can run normally under the operating system chosen by your company. For example, whether it supports WIN2000, Linux, etc.
· Development language: Is this software package suitable for the development language chosen by your company? For example, is it easy to use VC ++, VB, CPB, DELPHI and other languages for secondary development?
Of course, we have to compare the software packages of different companies in terms of the advantages and disadvantages of image processing functions.
· Edge search function: Edge search is the most basic and commonly used tool in image processing. This function mainly compares the following parameters: 1. Accuracy. You can consult the product introduction or instructions for the software package; 2. Stability. Different images can be used for actual testing of the software, such as GR & R testing; 3. Diversity. Whether it can cope with different practical situations. For example, edge search of different shapes, edge search under different light and dark conditions, etc .; 4. Speed. It is possible to test different systems in a unified situation (for example, using the same image).
· Target positioning function: Target positioning function is another basic and commonly used tool in image processing. There are also the following parameters for comparison: 1. Accuracy. You can consult the product introduction or instructions for the software package; 2. Stability. Different images can be used for actual testing of the software; 3. Speed. For the target positioning function, the speed of its running speed can be regarded as one of the direct signs of its software level. Under different circumstances (for example, using the same image and the same speed computer), different systems can be tested in practice; Fourth, the operation complexity. Since the target positioning function is much more complicated in algorithm than the edge search function, this makes the complexity of parameter setting of this function directly affect the difficulty of secondary development.
· Image pre-processing functions: image pre-processing functions (eg, binarization, edge sharpening, contrast adjustment, etc.) can improve image analysis speed and simplify analysis engineering.
· Character reading function (OCR): This function is especially important for visual systems that are mainly used for reading various characters.
· Data reading function: At present, most of the company's software packages have bar code and QR code reading functions. What everyone should pay attention to is the ability of different software to cope with unfavorable conditions, such as insufficient light source, unclear image, incomplete source code, etc.
· Image buffer function: similar to the "image storage" function of the board, the main purpose of this function is to alleviate the contradiction between taking pictures and computing time is not synchronized. Unlike the hardware buffer on the board, this function is implemented purely by software. The general concept is to create a fixed space in the memory, and the image signals from the board will be stored in this space in sequence before the computer analyzes it. The pointer of each image address is stored in another stack. The image is cleared in a first-in first-out manner.
· Practical application functions: In order to adapt to the increasingly fierce market competition, many companies have developed a software package that can be applied to a specific field on the basis of the original underlying software to facilitate visual secondary in this industry Developer. For example, BGA inspection tools in the semiconductor industry; display inspection tools in the display production industry; system positioning tools in the robot industry, etc.
· Interface function: Whether the software package can be easily interfaced with other software or controls to run together.
· Other auxiliary functions: In addition to the above-mentioned functions. Each vision company has also developed some other non-image processing software tools to facilitate the use of developers. For example, data analysis tools (SPC), image display tools, drawing tools, etc. Before choosing a camera with an asynchronous signal function, first, make sure that the board also has this function. This function is essential for systems that require extremely high operating speeds. Taking RS170 as an example, if there is no such function, the system waits for a photo to take 0 to 40 milliseconds. Second, progressive scan. More and more systems will use progressive scanning cameras, so we must first know whether the board supports this camera. Third, multi-resolution support. Not every board can support cameras of all resolutions. Fourth, multiple channels. Most boards can control more than two cameras at the same time. As for other uncommon functions, such as clock function, etc.