TWI579557B - Image detection method for printed substrate - Google Patents
Image detection method for printed substrate Download PDFInfo
- Publication number
- TWI579557B TWI579557B TW104131006A TW104131006A TWI579557B TW I579557 B TWI579557 B TW I579557B TW 104131006 A TW104131006 A TW 104131006A TW 104131006 A TW104131006 A TW 104131006A TW I579557 B TWI579557 B TW I579557B
- Authority
- TW
- Taiwan
- Prior art keywords
- image
- substrate
- sample
- standard
- component
- Prior art date
Links
Description
本發明係有關一種印刷基板的影像檢測方法，尤指一種每一元件圖層獨立對位後形成一校正影像再進行檢測的印刷基板的影像檢測方法。 The present invention relates to an image detecting method for a printed substrate, and more particularly to an image detecting method for a printed substrate in which each component layer is independently aligned to form a corrected image and then detected.
印刷電路板目前被廣泛運用於各種電子產品的必要元件，其性能好壞決定該電子產品的品質良劣。印刷電路板的製造過程相當複雜且精密，需要應用大量的機械加工、化學蝕刻、表面處理等製程技術才得以完成。為了確保印刷電路板成品品質，在製造印刷電路板的後端製程會加入檢測的流程。除了電器特性的檢驗之外，利用人工目測或搭配光學放大檢測設備來進行印刷電路板的外觀檢驗。然而，因印刷電路板微型化趨勢造成目測檢驗難度增加，以及人工目測的主觀誤差，使用電腦影像檢測系統進行檢驗已經逐漸成為目前印刷電路板外觀檢驗的主流。電腦影像檢測系統常使用自動光學檢驗(Automatic Optical Inspection,AOI)或自動視覺檢驗(Automatic Visual Inspection,AVI)等技術，以提供高速精確的瑕疵檢驗結果，達到檢測一致性以及快速檢測的效果。一般來說，自動光學檢驗(AOI)技術會先儲存印刷電路板的範本影像(template image)，該範本影像將被視為能夠正確運作的印刷電路板。再利用一影像擷取手段獲得一樣品印刷電路板的樣品影像(sample image)。將該樣品影像與該範本影像利用一影像運算方法進行比對，以取得該樣品影像與該範本影像之間相異的部分，並認定為該樣品印刷電路板的瑕疵結構。 Printed circuit boards are currently widely used in the necessary components of various electronic products, and their performance determines the quality of the electronic products. The manufacturing process of printed circuit boards is quite complicated and precise, and requires a large number of process technologies such as machining, chemical etching, and surface treatment to be completed. In order to ensure the quality of printed circuit boards, the inspection process is added to the back-end manufacturing process of printed circuit boards. In addition to the inspection of the electrical characteristics, the visual inspection of the printed circuit board is performed by manual visual inspection or with an optical amplification detecting device. However, due to the trend of miniaturization of printed circuit boards, the difficulty of visual inspection and the subjective error of manual visual inspection, the use of computer image detection systems for inspection has gradually become the mainstream of printed circuit board appearance inspection. Computer image detection systems often use techniques such as Automatic Optical Inspection (AOI) or Automatic Visual Inspection (AVI) to provide high-speed and accurate flaw detection results to achieve consistent detection and rapid detection. In general, Automated Optical Inspection (AOI) technology stores a template image of a printed circuit board (template). Image), the template image will be considered a printed circuit board that works correctly. A sample image of a sample printed circuit board is obtained by an image capture method. The sample image is compared with the template image by an image computing method to obtain a different portion between the sample image and the template image, and is identified as the 瑕疵 structure of the sample printed circuit board.
然而，各類印刷電路板在製造的過程中，尤其是軟性電路板，縱使是同一批製造的印刷電路板之間存在些微的差異。印刷電路板上各電路元件，有時會發生尺寸脹縮或是位移，因而影響其他正常電路元件的位置或結構，例如金屬層與阻焊層的交界處。當經過自動光學檢驗(Automatic Optical Inspection,AOI)或自動視覺檢驗(Automatic Visual Inspection,AVI)技術，這些結構上的差異將會明顯地被放大，而使正常的電路元件被電腦影像檢測系統定義為瑕疵。 However, various types of printed circuit boards are in the process of manufacturing, especially flexible circuit boards, even though there are slight differences between printed circuit boards manufactured in the same batch. The circuit components on the printed circuit board sometimes undergo dimensional expansion or displacement, thus affecting the position or structure of other normal circuit components, such as the junction of the metal layer and the solder resist layer. When subjected to Automatic Optical Inspection (AOI) or Automatic Visual Inspection (AVI) technology, these structural differences will be significantly amplified, and normal circuit components are defined by computer image detection systems as defect.
本發明之主要目的，在於解決傳統利用光學影像檢測印刷電路板造成誤判的問題。 The main object of the present invention is to solve the problem of the conventional use of optical images to detect false circuit caused by printed circuit boards.
為達上述目的，本發明提供一種印刷基板的影像檢測方法，包含步驟有：a)取得一具有複數元件區域的標準基板影像以及對應每一該標準元件區域的複數標準元件圖層，該標準元件圖層包含有一位置資訊以及至少一標準影像參數；b)擷取一樣品基板的樣品基板初始影像，該樣品基板初始影像包含有複數各別對應該標準元件區域的樣品元件區域；c)分析該樣品基板初始影像，產生對應該些樣品元件區域的複數樣品元件圖層，並獲 得該樣品元件圖層的一位置資訊以及樣品影像參數；d)根據該標準元件圖層與該樣品元件圖層的位置資訊，使每一該樣品元件圖層獨立對位於對應的該標準元件圖層；e)合併該些樣品元件圖層，形成一樣品基板校正影像；以及f)疊合該樣品基板校正影像與該標準基板影像，比對該樣品基板校正影像上每個該樣品元件圖層的該樣品影像參數與該標準基板影像上每個對應的該標準元件圖層的該標準影像參數，以檢測該樣品基板的各樣品元件區域是否具有瑕疵。 To achieve the above object, the present invention provides an image detecting method for a printed substrate, comprising the steps of: a) obtaining a standard substrate image having a plurality of component regions and a plurality of standard component layers corresponding to each of the standard component regions, the standard component layer Include a positional information and at least one standard image parameter; b) capture a sample substrate initial image of the sample substrate, the sample substrate initial image includes a plurality of sample element regions corresponding to the standard component regions; c) analyzing the sample substrate The initial image, generating a plurality of sample component layers corresponding to the sample element regions, and obtaining Obtaining a position information of the sample component layer and the sample image parameter; d) according to the position information of the standard component layer and the sample component layer, each of the sample component layers is independently located in the corresponding standard component layer; e) merging The sample component layers form a sample substrate calibration image; and f) superimposing the sample substrate calibration image and the standard substrate image, and correcting the sample image parameter of each sample component layer on the sample substrate The standard image parameter of each corresponding standard component layer on the standard substrate image to detect whether each sample element region of the sample substrate has defects.
於本發明一實施例中，步驟a)更包含步驟有：擷取一標準基板的該標準基板影像；以及分析該標準基板影像，產生對應該些標準元件區域的該標準元件圖層，並獲得該標準元件圖層的該位置資訊以及該標準影像參數。 In an embodiment of the invention, the step a) further comprises the steps of: capturing the standard substrate image of a standard substrate; and analyzing the standard substrate image to generate the standard component layer corresponding to the standard component regions, and obtaining the The location information of the standard component layer and the standard image parameters.
於本發明一實施例中，擷取該標準基板的該標準基板影像或擷取該樣品基板的該樣品基板初始影像，係利用攝影或掃描。 In an embodiment of the invention, the standard substrate image of the standard substrate or the initial image of the sample substrate of the sample substrate is captured by using photography or scanning.
於本發明一實施例中，該標準元件區域為金屬電路層區域、保護層區域、通孔區域或阻焊層區域。 In an embodiment of the invention, the standard component region is a metal circuit layer region, a protective layer region, a via region or a solder mask region.
於本發明一實施例中，該樣品元件區域為金屬電路層區域、保護層區域、通孔區域或阻焊層區域。 In an embodiment of the invention, the sample element region is a metal circuit layer region, a protective layer region, a via region, or a solder resist region.
於本發明一實施例中，該標準影像參數為多原色元素數值，該多原色元素數值為RGB、CMYK、YUV、HSV、LAB、XYZ、HSI或YIQ色彩空間。 In an embodiment of the invention, the standard image parameter is a multi-primary element value, and the multi-primary element value is an RGB, CMYK, YUV, HSV, LAB, XYZ, HSI or YIQ color space.
於本發明一實施例中，該樣品影像參數為多原色元素數值，該多原色元素數值為RGB、CMYK、YUV、HSV、LAB、XYZ、HSI或YIQ 色彩空間。 In an embodiment of the invention, the image parameter of the sample is a multi-primary element value, and the multi-primary element value is RGB, CMYK, YUV, HSV, LAB, XYZ, HSI or YIQ. Color space.
於本發明一實施例中，該印刷基板的影像檢測方法，更包含步驟有：設定該標準基板影像一定位點資訊；設定該樣品基板初始影像一定位點資訊；依據該定位點資訊，使該樣品基板初始影像對位疊合於該標準基板影像。 In an embodiment of the present invention, the image detecting method of the printed substrate further includes the steps of: setting the standard substrate image to a positioning point information; setting the initial image of the sample substrate to an positioning point information; and according to the positioning point information, The initial image alignment of the sample substrate is superposed on the standard substrate image.
於本發明一實施例中，該印刷基板的影像檢測方法，更包含步驟有：設定該標準基板影像一定位點資訊；設定該樣品基板校正影像一定位點資訊；依據該定位點資訊，使該樣品基板校正影像對位疊合於該標準基板影像。 In an embodiment of the present invention, the image detecting method of the printed substrate further includes the steps of: setting the standard substrate image to a positioning point information; setting the sample substrate to correct the image and the positioning point information; and according to the positioning point information, The sample substrate correction image is superposed on the standard substrate image.
有別於傳統電腦影像檢測系統將整片樣品影像與範本影像直接對位後再進行影像檢測運算，本發明印刷基板的影像檢測方法，先將標準基板影像中的各電路元件區分出對應的標準元件圖層，當擷取樣品基板的樣品基板初始影像後，亦依照各電路元件區分出複數樣品元件圖層，待每個該樣品元件圖層與該標準元件圖層對位後合併形成一樣品基板校正影像後，再讓該樣品基板校正影像與該標準基板影像進行影像檢測運算。如此，有效減少因為單一電路元件的差異而影響其他正常電路元件被電腦影像檢測系統誤判的發生，以提升電腦影像檢測系統檢測準確度與信效度。 Different from the traditional computer image detection system, the whole sample image and the template image are directly aligned, and then the image detection operation is performed. The image detection method of the printed substrate of the present invention first distinguishes each circuit component in the standard substrate image from the corresponding standard. After drawing the initial image of the sample substrate of the sample substrate, the component layer is further divided according to each circuit component, and after each sample component layer is aligned with the standard component layer, a sample substrate correction image is formed. And performing the image detection operation on the sample substrate calibration image and the standard substrate image. In this way, the occurrence of misjudgment of other normal circuit components by the computer image detection system is effectively reduced due to the difference of single circuit components, so as to improve the detection accuracy and reliability and validity of the computer image detection system.
10‧‧‧標準基板影像 10‧‧‧Standard substrate image
11‧‧‧金屬電路層區域 11‧‧‧Metal circuit layer area
12‧‧‧阻焊層區域 12‧‧‧ solder mask area
21‧‧‧標準元件圖層 21‧‧‧Standard component layers
22‧‧‧標準元件圖層 22‧‧‧Standard component layers
30‧‧‧樣品基板初始影像 30‧‧‧sample substrate initial image
31‧‧‧金屬電路層區域 31‧‧‧Metal circuit layer area
32‧‧‧阻焊層區域 32‧‧‧ Solder Mask Area
33、33’‧‧‧瑕疵 33, 33’‧‧‧瑕疵
41、41’‧‧‧樣品元件圖層 41, 41'‧‧‧ sample component layers
42、42’‧‧‧樣品元件圖層 42, 42'‧‧‧ sample component layers
50‧‧‧樣品基板校正影像 50‧‧‧Sample substrate calibration image
S10、S11、S12、S20、S30、S40、S50、S60、S71、S72、S73、S74、S75‧‧‧步驟 S10, S11, S12, S20, S30, S40, S50, S60, S71, S72, S73, S74, S75‧‧
圖1，為本發明印刷基板的影像檢測方法一實施例之主要步驟流程示意圖。 FIG. 1 is a flow chart showing the main steps of an embodiment of an image detecting method for a printed circuit board according to the present invention.
圖2，係本發明印刷基板的影像檢測方法另一實施例之步驟 流程示意圖。 2 is a view showing the steps of another embodiment of the image detecting method of the printed substrate of the present invention Schematic diagram of the process.
圖3-1至圖3-6，係本發明印刷基板的影像檢測方法一實施例之操作步驟示意圖。 3-1 to 3-6 are schematic diagrams showing the operation steps of an embodiment of the image detecting method of the printed circuit board of the present invention.
有關本發明之詳細說明及技術內容，現就配合圖式說明如下：請參閱『圖1』，本發明印刷基板的影像檢測方法一實施例之主要步驟流程示意圖，如圖所示：本發明印刷基板的影像檢測方法，包含步驟有：a)取得一具有複數元件區域的標準基板影像以及對應每一該標準元件區域的複數標準元件圖層，該標準元件圖層包含有一位置資訊以及至少一標準影像參數(步驟S10)；其中，該標準元件區域為金屬電路層區域、保護層區域、通孔區域或阻焊層區域；該標準影像參數為多原色元素數值，該多原色元素數值為RGB、CMYK、YUV、HSV、LAB、XYZ、HSI或YIQ色彩空間；b)擷取一樣品基板的樣品基板初始影像，該樣品基板初始影像包含有複數各別對應該標準元件區域的樣品元件區域(步驟S20)；其中，擷取該樣品基板的該樣品基板影像，係利用攝影或掃描；該樣品元件區域為金屬電路層區域、保護層區域、通孔區域或阻焊層區域；c)分析該樣品基板初始影像，產生對應該些樣品元件區域的複數樣品元件圖層，並獲得該樣品元件圖層的一位置資訊以及樣品影像參數(步驟S30)；其中，該樣品影像參數為多原色元素數值，該多原色元素數 值為RGB、CMYK、YUV、HSV、LAB、XYZ、HSI或YIQ色彩空間；d)根據該標準元件圖層與該樣品元件圖層的位置資訊，使每一該樣品元件圖層獨立對位於對應的該標準元件圖層(步驟S40)；e)合併該些樣品元件圖層，形成一樣品基板校正影像(步驟S50)；以及f)疊合該樣品基板校正影像與該標準基板影像，比對該樣品基板校正影像上每個該樣品元件圖層的該樣品影像參數與該標準基板影像上每個對應的該標準元件圖層的該標準影像參數，以檢測該樣品基板的各樣品元件區域是否具有瑕疵(步驟S60)。 The detailed description and technical contents of the present invention will now be described with reference to the following drawings: Referring to FIG. 1 , a schematic diagram of the main steps of an embodiment of the image detecting method for a printed circuit board of the present invention is shown in the figure: The image detecting method of the substrate comprises the steps of: a) obtaining a standard substrate image having a plurality of component regions and a plurality of standard component layers corresponding to each of the standard component regions, the standard component layer including a position information and at least one standard image parameter. (Step S10); wherein the standard component region is a metal circuit layer region, a protective layer region, a via region or a solder mask region; the standard image parameter is a multi-primary element value, and the multi-primary element value is RGB, CMYK, YUV, HSV, LAB, XYZ, HSI or YIQ color space; b) sampling a sample substrate initial image of the sample substrate, the sample substrate initial image containing a plurality of sample element regions corresponding to the standard component regions (step S20) Wherein the image of the sample substrate of the sample substrate is captured or photographed; the sample element region is a metal circuit layer region, a protective layer region, a via region or a solder mask region; c) analyzing the initial image of the sample substrate, generating a plurality of sample component layers corresponding to the sample component regions, and obtaining a position information of the sample component layer And sample image parameters (step S30); wherein the sample image parameters are multi-primary element values, and the number of the multi-primary elements Value is RGB, CMYK, YUV, HSV, LAB, XYZ, HSI or YIQ color space; d) according to the position information of the standard component layer and the sample component layer, so that each sample component layer is independently located corresponding to the standard a component layer (step S40); e) combining the sample component layers to form a sample substrate calibration image (step S50); and f) superimposing the sample substrate calibration image and the standard substrate image, and correcting the image to the sample substrate The sample image parameters of each of the sample element layers and the standard image parameters of the standard component layer corresponding to each of the standard substrate images are detected to detect whether each sample element region of the sample substrate has defects (step S60).
於一實施例中，請參閱『圖2』所示，步驟a)取得該標準基板影像以及複數標準元件圖層(步驟S10)，更包含步驟有：擷取一標準基板的該標準基板影像(步驟S11)；以及分析該標準基板影像，產生對應該些標準元件區域的該標準元件圖層，並獲得該標準元件圖層的該位置資訊以及該標準影像參數(步驟S12)。其中，擷取該標準基板的該標準基板影像，係利用攝影或掃描。 In an embodiment, referring to FIG. 2, step a) obtains the standard substrate image and the plurality of standard component layers (step S10), and further includes: capturing the standard substrate image of a standard substrate (step S11); and analyzing the standard substrate image, generating the standard component layer corresponding to the standard component regions, and obtaining the location information of the standard component layer and the standard image parameter (step S12). The standard substrate image of the standard substrate is captured or scanned.
為了能讓該樣品基板初始影像在進行分析之前，或者讓該樣品基板校正影像在進行比對之前，先與該標準基板影像對位，如『圖2』所示，本發明印刷基板的影像檢測方法更包含步驟有：設定該標準基板影像一定位點資訊(步驟S71)；設定該樣品基板初始影像一定位點資訊(步驟S72)；依據該定位點資訊，使該樣品基板初始影像對位疊合於該標準基板影像(步驟S73)；設定該樣品基板校正影像一定位點資訊(步驟S74)；依據該定位點資訊，使該樣品基板校正影像對位疊合於該標準基板影像(步驟S75)。 In order to allow the initial image of the sample substrate to be analyzed before the sample substrate is corrected or compared with the standard substrate image, as shown in FIG. 2, the image detection of the printed substrate of the present invention is performed. The method further includes the steps of: setting the standard substrate image-positioning point information (step S71); setting the sample substrate initial image-positioning point information (step S72); and according to the positioning point information, the sample substrate initial image alignment Combining the standard substrate image (step S73); setting the sample substrate correction image-positioning point information (step S74); and, according to the positioning point information, superimposing the sample substrate correction image on the standard substrate image (step S75) ).
請參閱『圖2』及『圖3-1』至『圖3-7』，係本發明印刷基板的影像檢測方法一實施例之操作步驟示意圖，如圖所示：在進行本發明檢測方法之前，須先選擇一無瑕疵且電路特性正常的印刷電路板，並將其定義為一標準基板。在進行樣品印刷電路板檢測之前，須要先設定每個電路元件圖層的標準位置資訊以及標準影像參數。首先，在步驟S11中，利用光學攝影或掃描等方式以擷取一標準基板的該標準基板影像10，如『圖3-1』所示，該標準基板影像10包含有具有複數元件區域；於本實施例中，該標準元件區域包含有金屬電路層區域11以及阻焊層區域12。擷取完該標準基板影像10後，設定該標準基板影像10的定位點資訊(步驟S71)。於步驟S12中，分析該標準基板影像10，產生對應該金屬電路層區域11以及該阻焊層區域12的標準元件圖層21、22，如『圖3-2』所示，並獲得該標準元件圖層21、22的位置資訊以及標準影像參數；於本實施例中，該標準影像參數為RGB色彩空間(RGB color space)，該金屬電路層區域11的標準元件圖層21內像素的RGB值為R：150~240，G：100~220；B：100~220，該阻焊層區域12的標準元件圖層22內像素的RGB值為R：20~40；G：80~150；B：20~50。 Please refer to FIG. 2 and FIG. 3-1 to FIG. 3-7, which are schematic diagrams showing the operation steps of an embodiment of the image detecting method of the printed circuit board of the present invention, as shown in the figure: before performing the detecting method of the present invention. A printed circuit board with no flaws and normal circuit characteristics must be selected and defined as a standard substrate. Before performing sample printed circuit board inspection, it is necessary to set the standard position information and standard image parameters of each circuit component layer. First, in step S11, the standard substrate image 10 of a standard substrate is captured by optical photography or scanning, as shown in FIG. 3-1, the standard substrate image 10 includes a plurality of component regions; In this embodiment, the standard component region includes a metal circuit layer region 11 and a solder resist layer region 12. After the standard substrate image 10 is captured, the positioning point information of the standard substrate image 10 is set (step S71). In step S12, the standard substrate image 10 is analyzed to generate standard component layers 21 and 22 corresponding to the metal circuit layer region 11 and the solder resist layer region 12, as shown in FIG. 3-2, and the standard component is obtained. The position information of the layers 21 and 22 and the standard image parameters. In this embodiment, the standard image parameter is an RGB color space, and the RGB value of the pixel in the standard component layer 21 of the metal circuit layer region 11 is R. : 150~240, G: 100~220; B: 100~220, the RGB value of the pixel in the standard component layer 22 of the solder mask region 12 is R: 20~40; G: 80~150; B: 20~ 50.
當定義該標準元件圖層21、22的位置資訊以及標準影像參數後，開始對一樣品基板進行影像處理。在步驟S20中，利用光學攝影或掃描等方式擷取一樣品基板的樣品基板初始影像30，如『圖3-3』所示，該樣品基板初始影像30包含有複數各別對應該標準元件區域的樣品元件區域；於本實施例中，該樣品元件區域亦包含有金屬電路層區域31以及阻焊層區域32。相較於該標準基板，該樣品基板的該阻焊層區域32與該金屬電路層區域31的交界處較高，也就是說該金屬電路層區域31露出區域較多。此外， 該金屬電路層區域31更具有至少一瑕疵33(以虛線圈起來部份)。擷取完該樣品基板初始影像30後，設定樣品基板初始影像30一定位點資訊(步驟S72)，並依據該定位點資訊，使該樣品基板初始影像30對位疊合於該標準基板影像10(步驟S73)。於步驟S30，分析該樣品基板初始影像30，產生對應金屬電路層區域31以及阻焊層區域32的樣品元件圖層41、42，如『圖3-4』所示，並獲得該樣品元件圖層41、42的一位置資訊以及樣品影像參數；於本實施例中，該樣品影像參數為RGB色彩空間(RGB color space)，該金屬電路層區域31的樣品元件圖層41內像素的RGB值為R：150~240，G：100~220；B：100~220，該阻焊層區域32的樣品元件圖層42內像素的RGB值為R：20~40；G：80~150；B：20~50。 After defining the position information of the standard component layers 21, 22 and the standard image parameters, image processing of a sample substrate is started. In step S20, a sample substrate initial image 30 of a sample substrate is captured by optical photography or scanning, as shown in FIG. 3-3, and the sample substrate initial image 30 includes a plurality of corresponding standard component regions. In the present embodiment, the sample element region also includes a metal circuit layer region 31 and a solder resist region 32. Compared with the standard substrate, the junction of the solder resist layer region 32 and the metal circuit layer region 31 of the sample substrate is higher, that is, the metal circuit layer region 31 has more exposed regions. In addition, The metal circuit layer region 31 further has at least one turn 33 (with a dotted portion). After the sample substrate initial image 30 is taken, the sample substrate initial image 30 is set to a positioning point information (step S72), and the sample substrate initial image 30 is aligned to the standard substrate image 10 according to the positioning point information. (Step S73). In step S30, the sample substrate initial image 30 is analyzed, and sample element layers 41 and 42 corresponding to the metal circuit layer region 31 and the solder resist layer region 32 are generated, as shown in FIG. 3-4, and the sample component layer 41 is obtained. In the embodiment, the image parameter of the sample is an RGB color space, and the RGB value of the pixel in the sample component layer 41 of the metal circuit layer region 31 is R: 150~240, G: 100~220; B: 100~220, the RGB values of the pixels in the sample component layer 42 of the solder resist region 32 are R: 20~40; G: 80~150; B: 20~50 .
如『圖3-5』所示，步驟S40說明根據該標準元件圖層21、22與該樣品元件圖層41、42的位置資訊，使每一該樣品元件圖層41、42獨立對位於對應的該標準元件圖層21、22。於本實施例中，該樣品基板的該阻焊層區域32的樣品元件圖層42沿著方向D位移，使其與該標準基板的該組焊層區域12的該標準元件圖層22完成對位。接著，請參考『圖3-6』所示，進行步驟S50，將獨立對位完成後的該樣品元件圖層41’、42’合併在一起，形成一樣品基板校正影像50。於該樣品基板校正影像50，該金屬電路層區域31的樣品元件圖層41’上方被該阻焊層區域32的樣品元件圖層42’遮蔽。形成該樣品基板校正影像50後，設定該樣品基板校正影像50的一定位點資訊(步驟S74)，並依據定位點資訊，使該樣品基板校正影像50(上層)對位疊合於該標準基板影像10(底層)(步驟S75)。最後，疊合該樣品基板校正影像50與該標準基板影像10，比對該樣品基板校正影像50上每個該樣品元件圖 層41’、42’的該樣品影像參數與該標準基板影像10上每個對應的該標準元件圖層21、22的該標準影像參數。該樣品基板校正影像50與該標準基板影像10上的差異，經校正後應排除無須比對的部份(如『圖3-6』中該金屬電路層區域31的樣品元件圖層41’上方被該阻焊層區域32的樣品元件圖層42’遮蔽的部份)，留下該金屬電路層區域31實際上的瑕疵33’。 As shown in FIG. 3-5, step S40 illustrates that the position information of the standard component layers 21, 22 and the sample component layers 41, 42 are such that each of the sample component layers 41, 42 is independently positioned corresponding to the standard. Component layers 21, 22. In the present embodiment, the sample element layer 42 of the solder resist layer region 32 of the sample substrate is displaced along the direction D to be aligned with the standard component layer 22 of the set of solder layer regions 12 of the standard substrate. Next, referring to Fig. 3-6, step S50 is performed to merge the sample element layers 41' and 42' after the independent alignment is completed to form a sample substrate correction image 50. The sample substrate correction image 50 is over the sample element layer 41' of the metal circuit layer region 31 and is shielded by the sample element layer 42' of the solder resist layer region 32. After the sample substrate calibration image 50 is formed, an anchor point information of the sample substrate calibration image 50 is set (step S74), and the sample substrate correction image 50 (upper layer) is aligned and aligned on the standard substrate according to the positioning point information. Image 10 (bottom layer) (step S75). Finally, the sample substrate calibration image 50 and the standard substrate image 10 are superimposed, and each sample component image on the sample substrate correction image 50 is compared. The sample image parameters of layers 41', 42' and the standard image parameters of the standard component layers 21, 22 corresponding to each of the standard substrate images 10. The difference between the sample substrate correction image 50 and the standard substrate image 10 is corrected, and the portion that does not need to be aligned is excluded (such as the sample element layer 41' of the metal circuit layer region 31 in FIG. 3-6. The portion of the solder mask region 32 that is masked by the sample element layer 42' leaves the actual 瑕疵33' of the metal circuit layer region 31.
本發明印刷基板的影像檢測方法，先將標準基板影像中的各電路元件區分出對應的標準元件圖層，當擷取樣品基板的樣品基板初始影像後，亦依照各電路元件區分出複數樣品元件圖層，待每個該樣品元件圖層與該標準元件圖層對位後合併形成一樣品基板校正影像後，再讓該樣品基板校正影像與該標準基板影像進行影像檢測運算。如此，有效減少因為單一電路元件的差異而影響其他正常電路元件被電腦影像檢測系統誤判的發生，以提升電腦影像檢測系統檢測準確度與信效度。 The image detecting method of the printed circuit board of the present invention first distinguishes each circuit component in the standard substrate image from the corresponding standard component layer, and after extracting the initial image of the sample substrate of the sample substrate, distinguishes the plurality of sample component layers according to each circuit component. After each of the sample component layers is aligned with the standard component layer to form a sample substrate calibration image, the sample substrate calibration image and the standard substrate image are subjected to image detection operations. In this way, the occurrence of misjudgment of other normal circuit components by the computer image detection system is effectively reduced due to the difference of single circuit components, so as to improve the detection accuracy and reliability and validity of the computer image detection system.
S10、S20、S30、S40、S50、S60‧‧‧步驟 S10, S20, S30, S40, S50, S60‧‧ steps
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW104131006A TWI579557B (en) | 2015-09-18 | 2015-09-18 | Image detection method for printed substrate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW104131006A TWI579557B (en) | 2015-09-18 | 2015-09-18 | Image detection method for printed substrate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201712326A TW201712326A (en) | 2017-04-01 |
| TWI579557B true TWI579557B (en) | 2017-04-21 |
Family
ID=59241149
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW104131006A TWI579557B (en) | 2015-09-18 | 2015-09-18 | Image detection method for printed substrate |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI579557B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI255338B (en) * | 2004-07-26 | 2006-05-21 | Dainippon Screen Mfg | Technique for detecting a defect of an object by area segmentation of a color image of the object |
| TW200949241A (en) * | 2008-05-28 | 2009-12-01 | Asustek Comp Inc | Apparatus and method for detecting circuit board |
| CN102138068B (en) * | 2008-09-09 | 2015-02-11 | 玛机统丽公司 | Visual examination apparatus |
| TW201522949A (en) * | 2013-12-03 | 2015-06-16 | Maintek Comp Suzhou Co Ltd | Inspection method for image data |
| CN103293170B (en) * | 2012-03-01 | 2015-08-19 | 斯克林集团公司 | Territorial classification device, base board checking device and territorial classification method |
-
2015
- 2015-09-18 TW TW104131006A patent/TWI579557B/en active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI255338B (en) * | 2004-07-26 | 2006-05-21 | Dainippon Screen Mfg | Technique for detecting a defect of an object by area segmentation of a color image of the object |
| TW200949241A (en) * | 2008-05-28 | 2009-12-01 | Asustek Comp Inc | Apparatus and method for detecting circuit board |
| CN102138068B (en) * | 2008-09-09 | 2015-02-11 | 玛机统丽公司 | Visual examination apparatus |
| CN103293170B (en) * | 2012-03-01 | 2015-08-19 | 斯克林集团公司 | Territorial classification device, base board checking device and territorial classification method |
| TW201522949A (en) * | 2013-12-03 | 2015-06-16 | Maintek Comp Suzhou Co Ltd | Inspection method for image data |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201712326A (en) | 2017-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4103921B2 (en) | Method of setting inspection reference data for fillet inspection, and substrate visual inspection apparatus using this method | |
| CN105136818B (en) | The image detection method of printed base plate | |
| JP5045591B2 (en) | Method for creating area setting data for inspection area and board appearance inspection apparatus | |
| JP2007155610A (en) | Visual examination device and visual examination method | |
| JP4454428B2 (en) | Component edge detection method, component edge detection program, and inspection apparatus | |
| JP4577717B2 (en) | Bump inspection apparatus and method | |
| KR100608225B1 (en) | Method for verification of the components of printed circuit board | |
| JP6165297B1 (en) | Substrate inspection apparatus and substrate manufacturing method | |
| TWI579557B (en) | Image detection method for printed substrate | |
| EP3104169B1 (en) | Quality management system | |
| TW201522949A (en) | Inspection method for image data | |
| JP4423130B2 (en) | Printed circuit board visual inspection method, printed circuit board visual inspection program, and printed circuit board visual inspection apparatus | |
| JP2006234554A (en) | Method and device for inspecting pattern | |
| JP2019045451A (en) | Inspection apparatus, inspection method and program | |
| JP2561193B2 (en) | Print pattern inspection device | |
| JP3857668B2 (en) | Pattern alignment method | |
| JP2006284543A (en) | Method and device for inspecting mounted circuit board | |
| TWI549097B (en) | A method of detecting images of appearance of electronic components and computer readable media thereof | |
| JP2010243214A (en) | Method and device for detection of flaw | |
| JP2008292342A (en) | Edge defect detecting method, program of the same, and detector | |
| JP2013011569A (en) | Displacement amount identification device, displacement amount identification method, and displacement amount identification program | |
| JP5096940B2 (en) | Inspection method and apparatus for printed wiring board | |
| JP2010276481A (en) | Image processing device | |
| JP2009145161A (en) | Method and apparatus for detecting defect | |
| JP2006329679A (en) | Method and device for inspecting product pattern |