Automatic optical inspection system is applied in PCBa production line after the SMD parts installed, or after solder, it automatically detects the placement status of parts and solder status, detects the bad pcb welding, Aoi is good helper for improving the finished product rate.

With the reduction of the device package size and the increase of circuit board patch density, SMD inspection is more and more difficult, manual visual inspection appears to be weak, its stability and reliability are difficult to meet the needs of production and quality control, so the use of special AOI inspection system to achieve automatic detection is more and more important. Optical instruments has a common feature, that is, through the light source of SMA reflected light acquisition, through the computer image processing system to determine the position of components on the SMA and welding, so this kind of equipment called automatic optical inspection (Automated Optical Inspection (AOI) .AOI optical inspection equipment is divided into two categories: online AOI and offline AOI systems. And AOI equipment is usually divided into three types according to the position on the assembly line: the first is the AOI placed in the screen printing to detect the failure of the solder paste, called the screen printing AOI; the second is called post-patch AOI; the third is placed after reflow welding or wave soldering to detect device placement and welding failures at the same time, called reflow welding AOI.

　The failure detection ability and principle of the above offline automated optical inspection equipment are described below. The AOI at this station can be used to detect various welding defects (including patch defects) after component welding. The test principle is that there is a relationship between the wetting angle between the welding strength WA and the solder/component pin, i.e. the welding strength of the solder can be inferred by the leaching state of the solder. The image captured by the camera, it is composed of many pixels, and each pixel has a certain grayscale value (0 to 255), so through a certain mathematical model of image information processing, welding quality can be accurately determined. Can detect missing parts, offsets, skeweds, standing, side stand, flipping, wrong parts, OCV, breakage, capacitive polarity reverse, tin, tin less, virtual welding, tin, copper foil pollution and so on.