CN101210889A - Holographic automatic optical detection system and method - Google Patents

Abstract

The invention discloses a holographic automatic optical detection system, which comprises a scanning device and an electronic computing device; the scanning device is provided with a driving unit and an optical scanning unit, and the optical scanning unit is driven by the driving unit to scan a full image of a printed circuit pattern; the electronic computing device obtains the picture file of the printed circuit pattern, converts the picture file into standard data which accords with a preset format, controls the driving unit to drive the optical scanning unit to scan the full image of the printed circuit pattern and converts the full image into data to be detected which accords with the preset format; then, the data to be tested is compared with the standard data to calibrate the data to be tested, so that the data to be tested and the standard data are aligned with each other, and the data to be tested is compared with the standard data to judge the defects of the printed circuit pattern.

Description

Holographic automatic optical inspection system and method

technical field

The present invention relates to an automatic optical inspection (Automated Optical Inspection; AOI) system and method, in particular to a holographic automatic optical inspection system and method.

Background technique

Referring to Fig. 1, the existing automatic optical inspection system 900 comprises a computer device 8 and an automatic optical inspection device 9, wherein the computer device 8 has an operation interface 81 and an input unit 82; the automatic optical inspection device 9 has a control unit 91 , a drive unit 92 , a transport unit 93 , an air extraction unit 94 , an image capture unit 95 and an illumination unit 96 .

Referring to Fig. 1 and Fig. 2, the operation method of the automatic optical inspection system 900 is to open an operation interface 81 for controlling the automatic optical inspection device 8 of the computer device 8, and first read the original circuit file of a circuit board by the operation interface 81 (step 701), and refer to the instruction of the operation interface 81 to input the setting value of the control signal through the input unit 82 (step 702).

Then start the actual detection. The detection step is that the user places a circuit board 7 to be tested on the top surface of the conveying unit 93, and the control unit 91 controls the driving unit 92 to drive the conveying unit 7 to move with the set control signal. Go to a position to be tested (step 703), and drive the suction unit 94 to pump air so that the circuit board 7 to be tested is positioned and kept flat (step 704).

After the circuit board 7 to be tested is transported to the position to be tested, the control unit 91 lights up the lighting unit 96 according to the set control signal to provide sufficient brightness for the circuit board 7 to be tested, and adjusts the height of the image capture unit 95 With clear focus, allow the image capture unit 95 to obtain the partial image of the circuit board 7 to the control unit 91 (step 705) every time it moves, and the control unit 91 outputs it to the computer device 9 for image reconstruction and the image of the circuit board 7 to be tested. Defect detection operation (step 706).

However, existing automated optical inspection systems 900 have the following disadvantages:

1. Each movement during testing can only capture a partial image of the circuit board 7 to be tested, and it takes time to complete the testing of all the images.

2. Online operation is adopted during testing, the circuit board 7 to be tested must be positioned and leveled with the air extraction unit 94, and the height of the image capture unit 95 needs to be automatically adjusted to focus clearly. The related control program of the circuit board 7 to be tested is complicated and difficult to operate.

3. The cost of the automatic optical inspection device 8 is expensive, and it can only perform random inspections, but cannot perform comprehensive inspections. Moreover, the mechanism of the automatic optical inspection device 8 is complex and bulky, and it needs to adopt an independent operation process, which cannot be matched with the existing maintenance system. Consistent work flow.

4. The defect detection operation includes a alignment program and a defect judgment program. The alignment program uses an existing component database. During detection, the data to be tested must be aligned with the components in the component database before subsequent defects can be performed. Judgment procedure, operation is very complicated.

Contents of the invention

In view of the fact that the current circuit board design tends to be more sophisticated, the requirements for quality control are becoming increasingly stringent. However, the cost of establishing an automatic optical inspection system is too high. In the absence of comprehensive inspection, the yield rate of electronic products will be reduced. Tested.

Therefore, the purpose of the present invention is to provide a holographic automatic optical inspection system and method with low cost, comprehensive defect detection, and a consistent operation process that can be matched with existing maintenance systems.

The holographic automatic optical detection system of the present invention detects a printed circuit pattern, and the printed circuit pattern is converted from an image file. The system includes a scanning device and an electronic computing device.

The scanning device has a driving unit and an optical scanning unit. The driving unit is controlled by a control signal to generate driving force. The optical scanning unit is driven by the driving unit to scan the hologram of the printed circuit pattern.

The electronic computing device has an input unit, a memory module and a control unit; the input module is operated by the user to generate a control signal; the receiving unit is used to receive the data to be tested obtained by the scanning device; the memory module is used to store the printed circuit diagram Such holograms and image files.

The control unit is used to obtain the picture file from the memory module, convert it into a standard data conforming to a predetermined format, and receive the control signal to control the driving unit to drive the optical scanning unit to scan the hologram of the printed circuit pattern, and convert it into conforming to the predetermined format The data to be tested in a format is temporarily stored in the memory module; then, the data to be tested is obtained from the memory module and compared with the standard data to calibrate the data to be tested, so that the data to be tested and the standard data are aligned with each other, and the data to be tested is The data is compared with the standard data to determine the defects of the printed circuit pattern.

The holographic automatic optical inspection method of the present invention detects a printed circuit pattern, and the printed circuit pattern is converted from an image file. The method includes the following steps: (A) converting the image file into a standard conforming to a predetermined format data; (B) scan the hologram of the printed circuit pattern, and convert it into a data to be tested in accordance with a predetermined format; (C) compare the data to be tested with the standard data to calibrate the data to be tested so that it is aligned with the standard data; and ( D) Compare the data to be tested with the standard data to determine the defects of the printed circuit pattern.

Since the holographic automatic optical inspection system and method of the present invention adopt a scanning device with high resolution to obtain the data to be tested, and cooperate with related automatic inspection steps, it can not only save costs, but also comprehensively detect defects, and is compatible with existing The maintenance system is combined into a consistent operation process, which can greatly improve the yield rate of the current circuit board manufacturing process for the industry.

Description of drawings

Below in conjunction with accompanying drawing and embodiment the present invention is described in detail:

Fig. 1 is a system block diagram illustrating that an existing automatic optical inspection system includes a computer device and an automatic optical inspection device;

Fig. 2 is a flow chart illustrating the existing automatic optical inspection method;

Fig. 3 is a system block diagram illustrating a preferred embodiment of the holographic automatic optical inspection system of the present invention;

Fig. 4 is a system block diagram illustrating another preferred embodiment of the holographic automatic optical inspection system of the present invention;

Fig. 5 is a flow chart illustrating a preferred embodiment of the holographic automatic optical inspection method of the present invention;

Figure 6 is a flowchart illustrating the sub-steps of step 46 of Figure 5;

FIG. 7 is a flowchart illustrating step 47 of FIG. 5 having sub-steps.

Detailed ways

The aforementioned and other technical contents, features and functions of the present invention will be clearly presented in the following detailed description of two preferred embodiments with reference to the drawings. Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same reference numerals.

Referring to Fig. 3, in a preferred embodiment of the present invention, the holographic automatic optical inspection system 100 includes a scanning device 1 and an electronic computing device 2, and the holographic automatic optical inspection system 100 is mainly for a printed circuit (PCB) ) pattern for detection, the printed circuit pattern is converted from a picture file and printed on a substrate, such as a circuit board 3, the picture file can be a negative file (Gerber file), or a shape with electronic components or circuits , position, etc. described computer graphics (CAD) files; however, it must be noted that the concept of the present invention is not limited to hard circuit board 3, as long as it is a plane with a printed circuit pattern, other substrates such as: soft board , film or film are also acceptable.

The scanning device 1 has an optical scanning unit 11, a driving unit 12, an upper cover 13 and a scanning platform 14; the circuit board 3 is placed on the scanning platform 14, and one side of the printed circuit pattern faces the optical scanning unit 11, and The downward pressure from the upper cover 13 makes the circuit board 3 positioned on the scanning platform 14 and becomes relatively flat.

The optical scanning unit 11 is driven by the driving unit 12 to scan the entire image of the circuit board 3, and the driving unit 12 is controlled by a control signal 101 sent from the electronic computing device 2 to generate a driving force to drive the optical scanning unit 11 Next, the optical scanning unit 11 scans the printed circuit pattern of the circuit board 3 to obtain a full image 201 and transmits the full image 201 to the electronic computing device 2 .

It must be noted that since the image must be finely identified, the scanning device 1 needs to match the minimum line width (line spacing) = x of the printed circuit pattern and select a resolution of x/4 to x/10 or above x/10 .

The electronic computing device 2 has a control unit 21 , a memory module 22 , a receiving unit 23 , a display unit 24 and an input module 25 .

The display unit 24 is used to display an operation interface 241, the user can refer to the options or instructions of the operation interface 241, and input the relevant control parameters through the input module 25 (such as: keyboard) to generate the control signal 101, the control parameters include Related settings such as scanning resolution, scanning range and image format; the receiving unit 12 is used to obtain the full image 201 from the scanning device 1; the memory module 24 is used to store the full image 201 and a Gerber file.

After the control unit 23 receives the control signal 101, it controls the scanning device 1 to start scanning with the control signal 101. After the scan is completed, the receiving unit 23 receives the full image 201 and converts it into a predetermined format (such as: ASCII code, but not in This is the limit) a data to be tested is temporarily stored in the memory module 22.

Referring to Fig. 4, in order to allow the holographic automatic optical inspection system 100 to carry out the detection work of multiple circuit boards, in another preferred embodiment, the holographic automatic optical inspection system 100 can also include a feeding device 15 and A receiving device 16, the feeding device 15 is loaded with a plurality of circuit boards with printed circuit patterns (shown at the end of the figure), and is used to accept the control of the electronic computing device 2' and automatically scan the scanning device 1 to obtain each circuit board The data to be tested; the receiving device is to collect the circuit board scanned by the scanning device 1 .

Referring to Fig. 3 and Fig. 5, in the preferred embodiment of the holographic automatic optical detection method of the present invention, when the user wants to analyze the printed circuit pattern defect of the circuit board 3, it is to open a set in the operation interface 241 An old file of control parameters (step 41) or a new file is opened (step 42). If a new file is opened, relevant control parameters must be set in the new file (step 43).

Then, open an image file of the printed circuit pattern on the operation interface 241 (step 44), and at this time, the control unit 23 obtains the image file from the memory module 22 and converts it into a standard data in a predetermined format (step 45), and Obtain the data to be tested and the standard data from the memory module 22 and compare them to calibrate the data to be tested (step 46), so that the data to be tested and the standard data are aligned; then, compare the data to be tested with the standard data to Defects of the printed circuit pattern are judged (step 47).

Referring to Fig. 6, the step 46 of Fig. 5 is illustrated to have following sub-steps: the full image 201 is image processed and converted into a data to be tested conforming to a predetermined format (step 461); each element is marked with feature extraction in the data to be tested Or line (step 462), its feature extraction can be to adopt the mode of morphology (morphology), topology (topology) or color analysis (colorimetric); and the rotation amount to adjust the data to be tested (step 463 ), so that the data to be tested and the standard data can be aligned with each other.

Among them, feature extraction refers to extracting the characteristics of each component or line of the data to be tested, such as: the shape, length, width or X, Y coordinate position of each component, etc., because the image file can be a negative file (Gerber file ), and the film file is an ASCII text file, which records the length, width, color or X, Y coordinate position of each component in the file, so it only needs to select the specific characteristics of each component to be tested data to align with standard data.

Referring to FIG. 7 , step 47 of FIG. 5 includes the following sub-steps: superimposing the data to be tested and the standard data (step 471 ); and marking the non-overlapping defects of the printed circuit patterns (step 472 ).

In this preferred embodiment, the non-overlapping defects include black defects (the missing part of the printed circuit pattern compared with the standard data) and white defects (the printed circuit pattern has more data than the standard data) Part of the above), the overlapping operation can be a subtraction operation between the data to be tested and the standard data, in addition to classification by morphological operations, edge operations can also be added for further classification; finally, the results of the classification operations are output Generate a defect analysis report (step 473).

To sum up the above, the holographic automatic optical inspection system and method of the present invention have the following advantages:

1. Since the scanning device 1 acquires all the images of the printed circuit pattern, the detection of all the data can be completed at one time during detection, which is quite time-saving.

2. Off-line operation is adopted during the detection, and there is no need to control the lighting of the lighting unit, the suction of the air extraction unit, the adjustment of the height of the image capture unit and other complicated setting procedures, which is very convenient for operation.

3. The cost of the scanning device 1 is low, it only needs sufficient resolution to perform comprehensive inspection, and it can be matched with the existing maintenance system to form a consistent operation process, so that the yield rate of the manufacturing process can be improved.

4. There is no need for a built-in component database for defect detection calculations. During detection, it is directly aligned with the components in the drawing, and the calculation is relatively fast.

Claims (13)

1. holographic type automatic optical detection system, be that a P.e.c. pattern is detected, described P.e.c. pattern is changed by figure shelves, described system comprises a computing electronics, has an input module, operated by the user and produce a controlling signal, it is characterized in that: described system also comprises:

The one scan device has:

One driver element is subjected to described controlling signal control to produce driving force; And

One light scanning unit is subjected to described drive unit drives to scan the full figure of described P.e.c. pattern;

Described computing electronics also has:

One receiving element is in order to receive the data to be measured that described scanister is obtained;

One memory module is in order to full figure and the described figure shelves that store described P.e.c. pattern; And

One control module, in order to obtain described figure shelves from described memory module, be converted into a standard data that meets a predetermined format, and receive described controlling signal and scan the full figure of described P.e.c. pattern to control the described light scanning unit of described drive unit drives, and be converted to meet described predetermined format a special survey data to be temporary in described memory module; Then, obtain described data to be measured and described standard data is compared to calibrate described data to be measured from described memory module, make described data to be measured align mutually, and described data to be measured and described standard data are compared to judge the defective of described P.e.c. pattern with described standard data.

2. holographic type automatic optical detection system as claimed in claim 1 is characterized in that:

Described scanister has more a loam cake and one scan platform, and described loam cake is to exert pressure so that a base material that is printed with described P.e.c. pattern entirely is positioned on the described scanning platform.

3. holographic type automatic optical detection system as claimed in claim 1 is characterized in that:

Described holographic type automatic optical detection system more comprises a feed arrangement and a material collecting device, and described feed arrangement and described material collecting device are all accepted described computing electronics control; Described feed arrangement is to load the most sheet circuit boards with described P.e.c. pattern, and controlledly give described scanister automatically and scan obtaining the data to be measured of each described circuit board, and described material collecting device is the described circuit board of collecting after the described scanister scanning.

4. holographic type automatic optical detection system as claimed in claim 1 is characterized in that:

Described scanister is to cooperate minimum feature (the line-spacing)=x of described P.e.c. pattern and select for use in x/4 to x/10 or the resolution more than the x/10.

5. holographic type automatic optical detection system as claimed in claim 1 is characterized in that:

The calibration of described control module is that described data to be measured is indicated each element with feature extraction, and adjusts data to be measured to align mutually with described standard data according to the relative displacement and the rotation amount of feature that data to be measured indicates.

6. holographic type automatic optical detection system as claimed in claim 5 is characterized in that:

The feature extraction of described control module is to adopt the mode of kenel, topology or color analysis to extract the feature of described each element of data to be measured or circuit.

7. holographic type automatic optical detection system as claimed in claim 1 is characterized in that:

The defective of described control module judges it is to indicate the underlapped defective of described P.e.c. pattern with described data to be measured and the two the overlapping computing of described standard data

8. holographic type automatic optical detection system as claimed in claim 7 is characterized in that:

Described underlapped defective comprises black defective and white defective.

9. a full image type automatic optical detection method is that a P.e.c. pattern is detected, and described P.e.c. pattern is changed by figure shelves, and it is characterized in that: described method comprises the steps:

(A) conversion of described figure shelves is met the standard data of a predetermined format;

(B) scanning described P.e.c. pattern full figure and be converted to a data to be measured that meets described predetermined format;

(C) the described data to be measured of comparison makes it align mutually with described standard data with described standard data to calibrate described data to be measured; And

(D) described data to be measured of comparison and described standard data are to judge the defective of described P.e.c. pattern.

10. full image type automatic optical detection method as claimed in claim 9 is characterized in that: step (C) comprises following substep:

(C-1) indicate each element in described data to be measured with feature extraction; And

(C-2) adjust data to be measured to align mutually according to the relative displacement and the rotation amount of feature that testing data indicates with described standard data.

11. full image type automatic optical detection method as claimed in claim 10 is characterized in that:

Feature extraction is to adopt the mode of kenel, topology or color analysis to extract the feature of described each element of data to be measured or circuit.

12. full image type automatic optical detection method as claimed in claim 9 is characterized in that: step (D) comprises following substep:

(D-1) the two is overlapping with described data to be measured and described standard data; And

(D-2) indicate the underlapped defective of described P.e.c. pattern.

13. full image type automatic optical detection method as claimed in claim 12 is characterized in that:

Described underlapped defective comprises black defective and white defective.

Images