Camera is an important part of machine vision. In machine vision, cameras are often used in industrial fields. Therefore, the choice of cameras in machine vision is often called the choice of industrial cameras. The quality of industrial camera selection often determines whether we can get the expected photos, so that the software can make accurate judgments. At the same time, in most application scenarios, the most suitable camera must be selected according to the user's needs, and the enterprise that formulates the solution should make the right choice for the camera in the project, and will also take the cost into account.
1）Signal types -- analog cameras and digital cameras
The output signal of analog camera is standard analog video signal. It needs special image acquisition card to convert analog signal into digital signal that can be processed by computer, so that the computer can process and apply the video signal later. Analog cameras have been widely used in early machine vision systems. They are suitable for machine vision systems with low image quality requirements.
The output signal of digital camera is digital signal. A/D conversion circuit is integrated in the camera, which directly converts the analog image signal into digital signal. It has the characteristics of strong anti-interference ability of image transmission, diverse video signal formats, high resolution and rich video output interface.
Machine vision is actually the processing of digital information in images. Analog cameras need to cooperate with image acquisition cards to convert images into digital information. Digital cameras do not need this link. Digital cameras have a higher dynamic range and can transmit more accurate signals to computers.
2）Chip type -- CCD and CMOS
CCD：charge coupled device
CMOS：Complementary Metal Oxide Semiconductor
3）Color：black and white camera and color camera
As the name implies, the color of the image acquired by the camera determines the color of the camera. Black-and-white cameras and color cameras have a wide range of applications. When the color features of the measured object are distinguished in disorder, black-and-white cameras can be used for detection. If the color becomes the key detection feature of the object to be measured, it needs to be used. Color cameras are used for testing.
4）Linear array camera and area array camera
As the name implies, linear array camera only gets one line of the object to be measured at a time, and scans the information of the direction of the object's movement through the continuous motion of the object; while the area array camera obtains a whole picture at one time, just like the mobile phone camera we commonly use. Therefore, the area array camera is generally used in the case of 2D detection, and the object to be measured is in discontinuous motion. Linear array camera can be used in 2D detection and 3D detection. In 2D detection, it is often used in the case of continuous application of the object to be measured. Linear array camera in 3D detection generally has depth of field characteristics, so it can obtain the depth of field. Height information of objects.
Common industrial camera interfaces are Camera Link, USB, Gige (Gigabit Network), 1394. Detailed parameter comparison can refer to the interface selection and comparison of industrial cameras.
2）Frame rate and line frequency
Frame rate refers to the number of pictures that a plane array camera can get per second, while line frequency refers to the line array camera, which refers to the line scan speed of the linear array camera.
The resolution of industrial cameras refers to the resolution of horizontal and vertical directions. We can usually calculate the roughly required resolution according to the accuracy requirements and the size of the object to be measured, and then choose the best camera.