JPH01209579A - Position recognizing device for ic - Google Patents

Abstract

PURPOSE:To highly accurately recognize the position of an IC even when the external shape of the IC becomes larger or it has more legs by dividing it with the use of plural ITV cameras and inputting IC pictures. CONSTITUTION:A flat package IC 1 is fixed on a transparent IC fixing stand 2, a light source 3 is arranged on the top face of the IC, and four ITV cameras 4 are arranged at the underside of the IC. Consequently, by successively taking in the four-divided pictures from the cameras and synthesizing them in a picture recognizing device 6, the whole image can be inputted, and the resolution of the camera is doubled compared with the case in which the whole is inputted at a time. Further, since the IC or the camera is not moved by a moving device, but the position relation of the respective cameras is fixed and simply switched by a camera selector 5, the position relation of the divided and inputted pictures is always constant, and the error due to division into the synthesized picture is not included. Thus, the position can be highly accurately recognized.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a position recognition device used in an IC mounter or an ICIAfdl inspection device.

[Summary of the Invention] The present invention is applicable to large-sized ICs in devices that require IC position recognition, such as IC mounters and IC visual inspection devices.
The position is recognized with high accuracy by inputting the information separately using a plurality of ITV cameras.

[Prior Art] Conventionally, the position of an IC has been recognized by inputting the entire IC using a single TV camera, or by using an XY
The IC is inputted in sections by fixing the IC to which moving device on the child table and moving the IC, or fixing the camera to the moving device and moving the camera.

[Problems to be solved by the invention] However, when the external size of the IC increases or the number of legs increases, there is a problem that a single ITV camera lacks resolution and cannot recognize the position.

When split input is performed by fixing an IC or a camera to a mobile device and moving it, there is a problem that the positioning error of the mobile device is added to the IC recognition position, making it impossible to accurately recognize the position.

In order to solve these conventional problems, the present invention aims to perform highly accurate position recognition without errors even when the external size of the IC increases or the number of legs increases.

[Means for Solving the Problems] The IC position recognition device of the present invention is characterized in that it uses a plurality of industrial TV cameras to perform position recognition by dividing and inputting the IC without moving either the IC or the camera.

[Function] Since division is performed using a plurality of ITV cameras, even if the resolution of each of the nine cameras is low, the resolution as a whole is high and highly accurate position recognition can be performed. Furthermore, since recognition is performed without moving either the input camera or the target IC, errors due to movement are not added.

[Examples] Examples of the present invention will be described below with reference to the drawings. In Figure 1, the flat package IC to be recognized
I is fixed to a transparent IC fixing base 2 made of glass. A light source 3 is placed on the top surface of the IC so as to sandwich the IC, and four ITV cameras 4 for inputting one image are placed below the IC. The fields of view of the cameras are installed so that the fields of view of the individual cameras do not overlap, as shown in FIG. 3, and so that the fields of view of the four cameras become one when combined. As shown in FIG. 2, the device inputs video signals from four industrial TV cameras 4 and a camera selector 5 that switches the cameras according to instructions from an image recognition device 6, and the video signal selected by the camera selector. The image recognition device 6 includes an image recognition device 6 that performs image recognition.

The images input by the four ITV cameras 4 are input in the form of a flat package ICI divided into four parts as shown in FIG. The entire image can be input by sequentially capturing the four divided images from the camera and combining them inside the image recognition device 6. Since the image is input divided into four parts and recombined within the image recognition device, the resolution of the camera is twice that of inputting the entire image at once. Also, by mobile device, IC
Alternatively, instead of moving the cameras, the positional relationship of the nine individual cameras is fixed and can only be selected by the camera selector 5, so the positional relationship of the divided and input images is always constant and the composite image This does not include errors due to division into .

[Effects of the Invention] As explained above, by using a plurality of ITV cameras and inputting the IC by dividing it, the position can be recognized with high accuracy even if the IC external shape is large or the number of ICs is large. This has the effect of allowing you to do this.

[Brief explanation of the drawing]

FIG. 1 is a side view of the input section of the present invention, and FIG. 2 is a side view of the input section of the present invention. The recognition device configuration diagram, FIG. 3, shows an input image of the IC. 1. Flat package IC 2. IC fixing stand 3. Light source 4. ITV camera 5. Camera selector 6. Image recognition device that's all Applicant: Seiko Epson Hayashi Company 7,'°sound Representative Patent Attorney Mogami: Tsutomu - 1 other person Figure 1 Figure 2

Claims (1)

[Claims] An IC position recognition device using an ITV camera as an input device, characterized in that a plurality of ITV cameras are used to divide and input an IC image.