Disclosure of Invention
The present invention is directed to a network centralized OCR recognition system and method, so as to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
a network centralized OCR recognition system comprises a high frame rate camera, a stroboscopic light source, a stepping motor and an OCR server, wherein the high frame rate camera is communicated with an embedded control chip and is used for acquiring an image of a wafer rotating synchronously with a turntable; the stepping motor is used for driving the turntable to rotate; the embedded control chip is used for sending the image obtained by the high-frame-rate camera and the angle and position information of the turntable to the OCR server, receiving the recognition result of the OCR server and controlling the corresponding action of the stepping motor to adjust the wafer; and the OCR server is used for identifying each frame of image obtained by the high frame rate camera and outputting an identification result.
As a still further scheme of the invention: the exposure time sequence of the high frame rate camera, the frequency of the stroboscopic light source and the rotation angle of the stepping motor are synchronous.
As a still further scheme of the invention: and the embedded control chip is communicated with the OCR server through a high-speed local area network.
As a still further scheme of the invention: the OCR server is connected with the embedded control chip through a local high-speed interface, and meanwhile, the OCR server is also connected with an upper computer through a communication interface.
As a still further scheme of the invention: the high frame rate cameras and the flash light sources are arranged in a plurality to form a camera module array, and the high frame rate cameras and the flash light sources are arranged on one side of the turntable according to the position sequence.
A network centralized OCR recognition method comprises the following steps:
s1, after grabbing the wafer, the mechanical arm is placed on a turntable driven by a stepping motor to rotate, and the wafer is driven by the turntable to rotate;
s2, the high frame rate camera shoots the wafer image under the action of the stroboscopic light source, and the exposure time sequence of the high frame rate camera is synchronous with the frequency of the stroboscopic light source;
s3, after the embedded control chip acquires the wafer image, the wafer image and the turntable angle position information when the current wafer image is shot are coded and sent to an OCR server, and the OCR server analyzes and identifies each frame of image;
and S4, when the key information point is identified, the OCR server outputs the turntable angle position information corresponding to the frame image to the embedded control chip, the embedded control chip controls the stepping motor to act to adjust the position, and after the adjustment is completed, the wafer is grabbed by the mechanical arm and transported back.
As a still further scheme of the invention: the high frame rate cameras and the flash light sources are all in multiple groups to form a camera module array, and the high frame rate cameras and the flash light sources are arranged on one side of the turntable according to the position sequence and output image information at different positions to the embedded control chip.
Compared with the prior art, the invention has the beneficial effects that: can be according to the user demand with the nimble arrangement of OCR module on local or remote industrial server, the computing power of make full use of current server need not specially designed or purchase the special module of making a video recording of special OCR, realizes good cost control.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Example 1
Referring to fig. 1 to 3, in an embodiment of the present invention, a network centralized OCR recognition system includes a high frame rate camera 100, a stroboscopic light source 200, an embedded control chip 300, a stepping motor 400, and an OCR server 500, where in this embodiment, the high frame rate camera 100 is in communication with the embedded control chip 300, and is configured to acquire an image of a wafer rotating synchronously with a turntable, and the stroboscopic light source 200 is in communication with the embedded control chip 300, and is configured to improve image acquisition quality of the high frame rate camera 100; the stepping motor 400 is communicated with the embedded control chip 300 and is used for driving the turntable to rotate; the OCR server 500 is in communication with the embedded control chip 300, and is configured to identify an image, that is, identify an edge and a key feature of a rotating wafer, and output an identification result.
Here, for each frame of image shot by the high frame rate camera 100, the embedded control chip 300 obtains the corresponding turntable angle position information, and encodes the two and outputs the two to the OCR server 500 (format: frame head + high frame rate camera ID + absolute angle + image data + frame end), where the turntable angle position information can be obtained by the control circuit of the stepping motor 400.
The OCR server 500 analyzes and recognizes each frame of image, when a key information point is recognized, the OCR server 500 returns the turntable angle position information corresponding to the frame of image to the embedded control chip 300, the embedded control chip 300 controls the stepping motor 400 to perform corresponding action, position adjustment is performed, after adjustment is performed, the wafer is grabbed and transported back by the mechanical arm, and because each frame of image and the turntable angle position information are synchronously coded or synchronized, network transmission delay does not affect recognition accuracy.
Specifically, the exposure timing of the high frame rate camera 100 and the frequency of the stroboscopic light source 200 are synchronized with the rotation angle of the stepping motor 400.
Specifically, the communication mode between the OCR server 500 and the embedded control chip 300 may be implemented as follows:
the embedded controller 300 communicates with the OCR server 500 through a high-speed lan, which is equivalent to arranging the OCR module in a local industrial server to form the OCR server 500, i.e. a remote recognition mode.
For a small-scale application scenario, the OCR module may be embedded in the embedded Linux platform to replace an industrial server, that is, actually, a combination of the OCR module and the embedded Linux platform may be placed at a local location of the device to implement offline independent operation, that is, the OCR server 500 is connected to the embedded control chip 300 through a local high-speed interface, and meanwhile, the OCR server 500 is further connected to an upper computer through a communication interface (a low-speed interface is preferably used as the communication interface, because it has a low requirement on communication speed, a common network is sufficient for use).
Meanwhile, in order to increase the operating speed of the entire system, the number of the high frame rate cameras 100 and the number of the stroboscopic light sources 200 are multiple, so as to form a camera module array, and the high frame rate cameras 100 and the stroboscopic light sources 200 are arranged on one side of the turntable according to the position sequence.
If the step angle of each frame of image is theta, N high frame rate cameras are in total, wherein the position of the first high frame rate camera is 0, the position of the second high frame rate camera is (A +1) theta, the position of the third high frame rate camera is (B +2) theta, the position of the nth high frame rate camera is (X + N) theta, wherein A, B … X is an integral multiple of N, and the installation is determined according to the specific situation, so that the whole system rotates at the speed of N times, and each camera module group simultaneously outputs image information at different positions.
Example 2
In the embodiment of the invention, a network centralized OCR recognition method comprises the following steps:
s1, after grabbing the wafer, the mechanical arm is placed on a turntable driven by a stepping motor to rotate, and the wafer is driven by the turntable to rotate;
s2, the high frame rate camera shoots the wafer image under the action of the stroboscopic light source, and the exposure time sequence of the high frame rate camera is synchronous with the frequency of the stroboscopic light source, so that the image quality can be improved, and later-stage identification is facilitated;
s3, after acquiring the wafer image, the embedded control chip encodes the wafer image and the turntable angle position information when the current wafer image is shot and sends the encoded wafer image and the turntable angle position information to an OCR server, and the OCR server analyzes and identifies each frame of image;
and S4, when the key information point is identified, the OCR server outputs the turntable angle position information corresponding to the frame image to the embedded control chip, the embedded control chip controls the stepping motor to act to adjust the position, and after the adjustment is completed, the wafer is grabbed by the mechanical arm and transported back.
Further, in order to improve the working efficiency, the high frame rate cameras and the stroboscopic light source are all in multiple groups to form a camera module array, and the high frame rate cameras 100 and the stroboscopic light source 200 are arranged on one side of the turntable according to the position sequence and output image information of different positions to the embedded control chip simultaneously without being repeated.
It should be noted that, in this technical solution, the OCR modules can be flexibly arranged on a local or remote industrial server according to the use requirements, the computing power of the existing server is fully utilized, the OCR module dedicated for shooting is not required to be specially designed or purchased, good cost control is realized, a plurality of OCR modules can be centrally arranged on a central server, scale application is realized, and the OCR modules run on a general computing platform, the realization threshold is low, development and debugging are not required for a dedicated chip, and the recognition mode can be flexibly changed and configured. In addition, the above embodiments only take the wafer edge finding as an example, and are taken as specific implementations of the present system.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.