本發明為有關一種偵測方法，尤指一種微型相機鏡頭影像瑕疵偵測方法。The invention relates to a detection method, in particular to a miniature camera lens image detection method.
鏡頭通常由一塊或多塊光學玻璃組合而成，理論上而言，一個簡單的凸透鏡就是一個鏡頭，但是在實際應用中，鏡頭需要各種透鏡的組合來矯正光學畸變，且隨著科技進步，除了利用各種凹凸透鏡組合來矯正光學畸變外，更會進一步對組合完成之鏡頭進行透鏡的瑕疵檢測，如刮傷、灰塵等，以確保品質。The lens is usually composed of one or more optical glasses. In theory, a simple convex lens is a lens, but in practical applications, the lens requires a combination of various lenses to correct optical distortion, and with the advancement of technology, in addition to In addition to correcting optical distortion by using various lenticular lens combinations, the lens of the combined lens is further detected, such as scratches, dust, etc., to ensure quality.
習知的檢測方式如王詩婷於2013年在國立臺北科技大學所發表之論文「應用機器視覺於微型照相模組之瑕疵檢測」，其為將待檢測影像分割成數個環狀區塊，再找出相對亮的瑕疵，但對於較暗、低對比的瑕疵則無法檢測出。The well-known detection method is the paper “Application of Machine Vision to the Detection of Micro Camera Modules” published by Wang Shiting at the National Taipei University of Science and Technology in 2013, which divides the image to be detected into several annular blocks and finds out Relatively bright, but not for darker, lower contrasting flaws.
且此種方法需要將待檢測影像分割成數個區塊，在判別瑕疵上，需要較長的時間，而不利於產業生產，因此，如何快速的檢知出瑕疵，實為一重要課題。Moreover, this method needs to divide the image to be detected into several blocks, and it takes a long time to determine the flaw, which is unfavorable for industrial production. Therefore, how to quickly detect the flaw is an important issue.
本發明的主要目的，在於解決無法快速的檢知出瑕疵的問題。The main object of the present invention is to solve the problem that the flaw cannot be quickly detected.
為達上述目的，本發明提供一種微型相機鏡頭影像瑕疵偵測方法，包含有以下步驟：S1：擷取一微型相機鏡頭之一鏡頭原始影像；S2：對該鏡頭原始影像進行極座標轉換，而形成一直角座標轉換影像；S3：利用一邊緣偵測演算技術來對該直角座標轉換影像進行邊緣偵測，以形成一瑕疵判別影像，並於該瑕疵判別影像上判斷一瑕疵；以及S4：將該直角座標轉換影像以及該瑕疵判別影像做元件連通標示，以確知該瑕疵位於該直角座標轉換影像上的形狀與位置。In order to achieve the above object, the present invention provides a micro camera lens image detection method, which comprises the following steps: S1: capturing a lens original image of a miniature camera lens; S2: performing polar coordinate conversion on the original image of the lens to form a continuous coordinate conversion image; S3: an edge detection algorithm is used to perform edge detection on the rectangular coordinate conversion image to form a discriminant image, and judge the image on the image; and S4: The rectangular coordinate conversion image and the 瑕疵 discrimination image are connected to the component to confirm the shape and position of the 瑕疵 on the orthogonal coordinate conversion image.
綜上所述，本發明利用極座標轉換而形成該直角座標轉換影像，再進行邊緣偵測，可以降低影像處理的複雜度，可直接進行檢測，免去對鏡頭原始影像進行切割的步驟，而可快速的檢知出瑕疵。In summary, the present invention utilizes polar coordinate conversion to form the rectangular coordinate conversion image, and then performs edge detection, which can reduce the complexity of image processing, and can directly perform detection, eliminating the step of cutting the original image of the lens, but Quickly detect the flaws.
S1~S4、S1A、S1B、S3A~S3D‧‧‧步驟S1~S4, S1A, S1B, S3A~S3D‧‧‧ steps
圖1，為本發明一較佳實施例的流程示意圖。FIG. 1 is a schematic flow chart of a preferred embodiment of the present invention.
圖2A~2F，為本發明一較佳實施例的檢測流程示意圖。2A-2F are schematic diagrams showing a detection process according to a preferred embodiment of the present invention.
有關本發明的詳細說明及技術內容，現就配合圖式說明如下：請參閱「圖1」至「圖2F」所示，本發明為一種微型相機鏡頭影像瑕疵偵測方法，於於其中，「圖2A」至「圖2F」為使用照片圖檔進行示意說明，部分照片為彩色，包含有以下步驟：The detailed description and the technical content of the present invention are as follows: Referring to FIG. 1 to FIG. 2F, the present invention is a micro camera lens image detection method, in which " Figures 2A” to 2F show the use of photo files. Some of the photos are colored and contain the following steps:
S1：如「圖2A」所示，先擷取一微型相機鏡頭之一鏡頭原始影像，而於此步驟後，更包含有以下步驟：S1: As shown in FIG. 2A, the original image of one of the miniature camera lenses is captured first, and after this step, the following steps are further included:
S1A：由於該微型相機鏡頭的遮光片外鏡邊緣為圓環狀邊緣，故該微型相機鏡頭具有一已知半徑，因此，可以透過霍夫轉換(hough transform)找出該鏡頭原始影像中相對於該已知半徑最多共圓的圓心座標，霍夫轉換為先考慮該鏡頭原始影像上的任一點(x,y)，圓心座標為(a,b)，r為半徑且過這一點的圓形方程式可以表示為：(x-a)2+(y-b)2=r2S1A: Since the outer lens edge of the micro camera lens has a ring-shaped edge, the micro camera lens has a known radius. Therefore, the hough transform can be used to find out the original image of the lens relative to The center coordinate of the known radius with the most common circle, the Hough transform is to consider any point (x, y) on the original image of the lens, the center of the circle is (a, b), r is the radius and the circle is over this point The equation can be expressed as: (xa)2 + (yb)2 = r2
再將歐基里德空間中的特徵點座標轉換映射至三維參數空間中，並將點座標(x,y)帶入公式轉換至參數空間中，在此半徑r為未知參數，故需自己設定r值的範圍，而因為圓偵測的原理跟直線偵測的原理是相似的，所以可得知在歐基里德空間中共圓的點座標，會在參數空間中相交於同一參數點座標上，以參數點(a,b,r)形式來表示，然而，因為半徑r不同，所以在參數空間中會以圓錐體形狀表示出來，同時，歐基里德空問中，一個座標點映射至參數空間中可以繪出一個對應的圓錐體，所以數個座標點經過轉換可以在參數空間中映射出跟座標點數目一樣多的圓錐體，因為參數空間中圓錐體的交點部分，就是歐基里德空間中共圓的部分，當參數點(a,b,r)累計次數達到所設的門檻值次數時，則將相對應的參數(a,b,r)回傳，即可得知最多共圓的圓心座標。Then map the feature point coordinates in the Euclid space to the 3D parameter space, and bring the point coordinates (x, y) into the formula to convert to the parameter space. Here, the radius r is an unknown parameter, so you need to set it yourself. The range of r values, and because the principle of circle detection is similar to the principle of line detection, it can be known that the coordinates of the coordinates of the circle in the Euclid space intersect at the coordinates of the same parameter point in the parameter space. , expressed in the form of parameter points (a, b, r), however, because the radius r is different, it will be represented by the shape of the cone in the parameter space. At the same time, in the Euclid space, a coordinate point is mapped to A corresponding cone can be drawn in the parameter space, so several coordinate points can be converted to map as many cones as the number of coordinate points in the parameter space, because the intersection of the cone in the parameter space is Oujiri. In the part of the circle in the German space, when the cumulative number of parameter points (a, b, r) reaches the set threshold number, the corresponding parameters (a, b, r) are returned, and the most common Round center coordinates.
S1B：得到最多共圓的圓心座標後，由於該微型相機鏡頭放置的位置可能會產生平移誤差，故對該鏡頭原始影像進行定位與擷取檢測區域，以減少平移誤差的發生。S1B: After obtaining the circle coordinate of the most common circle, since the position of the lens of the miniature camera may cause a translation error, the original image of the lens is positioned and captured to reduce the occurrence of translation error.
S2：如「圖2B」所示，進行霍夫轉換後，可得知出現頻率最高的圓心座標(a,b,r)，即可找出該鏡頭原始影像的圓心座標(a,b)，當得到圓心位置後，用影像內插法重新取樣，以對該鏡頭原始影像進行極座標轉換，而形成一直角座標轉換影像，即為將圓形的該鏡頭原始影像轉化為長條狀的該直角座標轉換影像，本實施例中，為清楚表示瑕疵部分，僅擷取了瑕疵部分的該直角座標轉換影像作為示意。S2: As shown in Fig. 2B, after performing Hough conversion, the center coordinates (a, b, r) with the highest frequency can be known, and the center coordinates (a, b) of the original image of the lens can be found. After obtaining the position of the center of the circle, re-sampling by image interpolation to perform polar coordinate conversion on the original image of the lens to form a rectangular coordinate conversion image, that is, converting the circular original image of the lens into a long stripIn the present embodiment, in order to clearly show the 瑕疵 portion, only the rectangular coordinate conversion image of the 瑕疵 portion is taken as an illustration.
S3：如「圖2C」至「圖2E」所示，經極座標轉換後，該鏡頭原始影像已變為水平直線特徵為主之該直角座標轉換影像，再利用邊緣偵測演算技術來對該直角座標轉換影像進行邊緣偵測，以形成一瑕疵判別影像，並於該瑕疵判別影像上判斷一瑕疵，其中，該邊緣偵測演算技術可以為邊緣強化偵測演算法(EEDA，Edge Enhancement Detection Algorithm)、梭柏演算法(Sobel operator)普魯伊特演算法(Prewitt operator)、坎尼演算法(Canny edge detector)、二值化演算法、辨識邊緣偵測演算法、本拉(MURA)邊緣偵測演算法、多尺度的邊緣偵測(multiscale edge detection)演算、座標旋轉數位計算(Coordinate Rotation Digital Computer，CORDIC)的遞迴演算法、雙向考量之多餘讀取器移除演算法(DOE)及蛇模型演算法(active contour model)等等，而於本實施例中，係使用梭柏演算法進行邊緣偵測。S3: As shown in Fig. 2C to Fig. 2E, after the polar coordinate conversion, the original image of the lens has become the horizontal coordinate feature of the rectangular coordinate conversion image, and then the edge detection algorithm is used to calculate the right angle. The coordinate conversion image is edge-detected to form a discriminating image, and the image is judged on the discriminating image, wherein the edge detection algorithm can be an Edge Enhancement Detection Algorithm (EEDA). , Sobel operator Prewitt operator, Canny edge detector, binarization algorithm, recognition edge detection algorithm, MURA edge detection Measurement algorithm, multiscale edge detection (multiscale edge detection), coordinate recursive digit computing (CORDIC) recursive algorithm, two-way consideration of redundant reader removal algorithm (DOE) and An active contour model or the like is used, and in the present embodiment, the Sorb algorithm is used for edge detection.
S3A：如「圖2C」及「圖2D」所示，對該直角座標轉換影像的每一個像素之周邊的八個像素分別做水平方向以及垂直方向的迴旋積計算，而可得到一相關於水平方向的水平梯度值以及一相關於垂直方向的垂直梯度值，其中，每一個像素及其周邊的八個像素可形成一九宮格，而位於周邊的八個像素分別定義如下表一所示：
而水平梯度值Gx=(Z3+2Z6+Z9)-(Z1+2Z4+Z7)、垂直梯度值Gy=(Z7+2Z8+Z9)-(Z1+2Z2+Z3)。The horizontal gradient value Gx=(Z3+2Z6+Z9)-(Z1+2Z4+Z7), the vertical gradient value Gy=(Z7+2Z8+Z9)-(Z1+2Z2+Z3).
而除了形成九宮格做迴旋積計算外，亦可以形成二十五宮格，即對該直角座標轉換影像的每一個像素之周邊的二十四個像素分別做水平方向以及垂直方向的迴旋積計算，同樣可以得到一相關於水平方向的水平梯度值以及一相關於垂直方向的垂直梯度值，而位於周邊的二十四個像素分別定義如下表二所示：
其中，水平梯度值Gx=(Z5+4Z10+6Z15+4Z20+Z25)+(2Z4+8Z9+12Z14+8Z19+2Z24)-(2Z2+8Z7+12Z12+8Z17+2Z22)-(Z1+4Z6+6Z11+4Z16+Z21)，而垂直梯度值Gy=(Z1+4Z2+6Z3+4Z4+Z5)+(2Z6+8Z7+12Z8+8Z9+2Z10)-(2Z16+8Z17+12Z18+8Z19+2Z20)-(Z21+4Z22+6Z23+4Z24+Z25)。Among them, the horizontal gradient value Gx=(Z5+4Z10+6Z15+4Z20+Z25)+(2Z4+8Z9+12Z14+8Z19+2Z24)-(2Z2+8Z7+12Z12+8Z17+2Z22)-(Z1+4Z6+6Z11+ 4Z16+Z21), and the vertical gradient value Gy=(Z1+4Z2+6Z3+4Z4+Z5)+(2Z6+8Z7+12Z8+8Z9+2Z10)-(2Z16+8Z17+12Z18+8Z19+2Z20)-(Z21+ 4Z22+6Z23+4Z24+Z25).
S3B：對該垂直梯度值進行加權計算，而得到一加權垂直梯度值，而該水平梯度值亦可以進行加權而得到一加權水平梯度值，於本實施例中，該水平梯度值為加權20%，該垂直梯度值為加權80%。舉例來說，上述梯度大小可分別加權改寫為加權水平梯度值Gmx=0.2Gx，加權垂直梯度值Gmy=0.8Gy。當然的，加權的比例可以依照實際需求進行調整，但於本實施例中，為了更輕易地取得瑕疵位置的所在，水平梯度值的加權必須小於垂直梯度值的加權，以凸顯垂直方向的缺陷問題。S3B: weighting the vertical gradient value to obtain a weighted vertical gradient value, and the horizontal gradient value may also be weighted to obtain a weighted horizontal gradient value. In this embodiment, the horizontal gradient value is weighted by 20%. The vertical gradient value is weighted by 80%. For example, the gradient magnitudes may be separately weighted to a weighted horizontal gradient value Gmx=0.2Gx, and the weighted vertical gradient value Gmy=0.8Gy. Of course, the weighted ratio can be adjusted according to actual needs, but in this embodiment, in order to more easily obtain the position of the 瑕疵 position, the weight of the horizontal gradient value must be smaller than the weight of the vertical gradient value to highlight the defect problem in the vertical direction. .
S3C：藉由該加權垂直梯度值得到該像素的一判別值，其主要方法可以分為以下兩種，一為將該加權垂直梯度值及該加權水平梯度值的絕對值相加，二為該加權垂直梯度值及該水平梯度值的絕對值相加，而得到該判別值，而於本實施例中，係進行該加權垂直梯度值與該加權水平梯度值的絕對值相加，並可利用下列公式作為取得絕對值的方式，藉此計算出梯度大小：
S3D：如「圖2E」所示，若該判別值小於一臨界值，可以使「圖2D」中較為細小而不存在的候選瑕疵消除，而若該判別值大於一臨界值，則判斷為瑕疵。S3D: As shown in Fig. 2E, if the discriminant value is less than a critical value, the candidate 瑕疵 which is smaller and does not exist in Fig. 2D can be eliminated, and if the discriminant value is greater than a critical value, it is judged as 瑕疵.
S4：最後如「圖2F」所示，將該直角座標轉換影像以及該瑕疵判別影像做元件連通標示，以確知並標示出該瑕疵位於該直角座標轉換影像上的形狀與位置。S4: Finally, as shown in FIG. 2F, the rectangular coordinate conversion image and the 瑕疵 discrimination image are connected to each other to identify and mark the shape and position of the cymbal on the rectangular coordinate conversion image.
綜上所述，本發明具有以下特點：In summary, the present invention has the following features:
一、利用極座標轉換而形成該直角座標轉換影像，再進行邊緣偵測，可以降低影像處理的複雜度，可直接進行檢測，免去對鏡頭原始影像進行切割的步驟，而可快速的檢知出瑕疵。1. Using the polar coordinate conversion to form the right-angle coordinate conversion image, and then performing edge detection, the complexity of the image processing can be reduced, and the detection can be directly performed, thereby eliminating the step of cutting the original image of the lens, and quickly detecting the result. defect.
二、藉由對該垂直梯度值進行加權計算，減少因鏡頭邊緣在水平方向的邊線被認定為瑕疵的狀況，而加強垂直方向瑕疵的偵測，因而可以更容易的判別出瑕疵。Second, by weighting the vertical gradient value, the edge of the lens edge in the horizontal direction is determined to be 瑕疵, and the detection of the vertical direction 加强 is enhanced, so that the 瑕疵 can be more easily discriminated.
因此本發明極具進步性及符合申請發明專利的要件，爰依法提出申請，祈 鈞局早日賜准專利，實感德便。Therefore, the present invention is highly progressive and conforms to the requirements of the invention patent application, and the application is filed according to law, and the praying office grants the patent as soon as possible.
以上已將本發明做一詳細說明，惟以上所述者，僅為本發明的一較佳實施例而已，當不能限定本發明實施的範圍。即凡依本發明申請範圍所作的均等變化與修飾等，皆應仍屬本發明的專利涵蓋範圍內。The present invention has been described in detail above, but the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the scope of the invention. That is, the equivalent changes and modifications made by the scope of the present application should remain within the scope of the patent of the present invention.
S1~S4、S1A、S1B、S3A~S3D‧‧‧步驟S1~S4, S1A, S1B, S3A~S3D‧‧‧ steps
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105133851ATWI606422B (en) | 2016-10-20 | 2016-10-20 | Miniature camera lens image detection method |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105133851ATWI606422B (en) | 2016-10-20 | 2016-10-20 | Miniature camera lens image detection method |
| Publication Number | Publication Date |
|---|---|
| TWI606422Btrue TWI606422B (en) | 2017-11-21 |
| TW201816717A TW201816717A (en) | 2018-05-01 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW105133851ATWI606422B (en) | 2016-10-20 | 2016-10-20 | Miniature camera lens image detection method |
| Country | Link |
|---|---|
| TW (1) | TWI606422B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11209795B2 (en) | 2019-02-28 | 2021-12-28 | Nanotronics Imaging, Inc. | Assembly error correction for assembly lines |
| US11156991B2 (en) | 2019-06-24 | 2021-10-26 | Nanotronics Imaging, Inc. | Predictive process control for a manufacturing process |
| US11063965B1 (en) | 2019-12-19 | 2021-07-13 | Nanotronics Imaging, Inc. | Dynamic monitoring and securing of factory processes, equipment and automated systems |
| US11100221B2 (en) | 2019-10-08 | 2021-08-24 | Nanotronics Imaging, Inc. | Dynamic monitoring and securing of factory processes, equipment and automated systems |
| TWI760916B (en) | 2019-11-06 | 2022-04-11 | 美商奈米創尼克影像公司 | Manufacturing system for automatic production line in factory |
| US12165353B2 (en) | 2019-11-06 | 2024-12-10 | Nanotronics Imaging, Inc. | Systems, methods, and media for manufacturing processes |
| US12153408B2 (en) | 2019-11-06 | 2024-11-26 | Nanotronics Imaging, Inc. | Systems, methods, and media for manufacturing processes |
| WO2021102223A1 (en) | 2019-11-20 | 2021-05-27 | Nanotronics Imaging, Inc. | Securing industrial production from sophisticated attacks |
| US11086988B1 (en) | 2020-02-28 | 2021-08-10 | Nanotronics Imaging, Inc. | Method, systems and apparatus for intelligently emulating factory control systems and simulating response data |
| WO2021183468A1 (en)* | 2020-03-09 | 2021-09-16 | Nanotronics Imaging, Inc. | Systems, methods, and media for manufacturing processes |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200725343A (en)* | 2005-12-29 | 2007-07-01 | Ind Tech Res Inst | Method and system for designing and manufacturing lens modules |
| CN101210890A (en)* | 2006-12-28 | 2008-07-02 | 夏普株式会社 | Defect detection device and method, image sensor device and module |
| US20080285840A1 (en)* | 2005-12-26 | 2008-11-20 | Nikon Corporation | Defect inspection apparatus performing defect inspection by image analysis |
| TW200849968A (en)* | 2007-06-12 | 2008-12-16 | Univ Nat Chunghsing | Image inpainting method for image defects caused by dusted camera sensor unit |
| CN101822039A (en)* | 2007-08-31 | 2010-09-01 | 奥多比公司 | Systems and methods for determination of camera imperfection for an image |
| TW201135668A (en)* | 2010-04-12 | 2011-10-16 | Univ Nat Taipei Technology | Image defect instpection system and method |
| TW201344168A (en)* | 2012-04-19 | 2013-11-01 | Benq Materials Corp | A contact lens detecting system |
| CN105044124A (en)* | 2015-08-27 | 2015-11-11 | 李明英 | Glass flaw classification device based on gray mean value analysis |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080285840A1 (en)* | 2005-12-26 | 2008-11-20 | Nikon Corporation | Defect inspection apparatus performing defect inspection by image analysis |
| TW200725343A (en)* | 2005-12-29 | 2007-07-01 | Ind Tech Res Inst | Method and system for designing and manufacturing lens modules |
| CN101210890A (en)* | 2006-12-28 | 2008-07-02 | 夏普株式会社 | Defect detection device and method, image sensor device and module |
| TW200849968A (en)* | 2007-06-12 | 2008-12-16 | Univ Nat Chunghsing | Image inpainting method for image defects caused by dusted camera sensor unit |
| CN101822039A (en)* | 2007-08-31 | 2010-09-01 | 奥多比公司 | Systems and methods for determination of camera imperfection for an image |
| TW201135668A (en)* | 2010-04-12 | 2011-10-16 | Univ Nat Taipei Technology | Image defect instpection system and method |
| TW201344168A (en)* | 2012-04-19 | 2013-11-01 | Benq Materials Corp | A contact lens detecting system |
| CN105044124A (en)* | 2015-08-27 | 2015-11-11 | 李明英 | Glass flaw classification device based on gray mean value analysis |
| Publication number | Publication date |
|---|---|
| TW201816717A (en) | 2018-05-01 |
| Publication | Publication Date | Title |
|---|---|---|
| TWI606422B (en) | Miniature camera lens image detection method | |
| CN107798326B (en) | A Contour Vision Detection Method | |
| CN107543828B (en) | Workpiece surface defect detection method and system | |
| Placht et al. | Rochade: Robust checkerboard advanced detection for camera calibration | |
| CN104112269B (en) | A kind of solar battery laser groove parameter detection method and system based on machine vision | |
| CN109900711A (en) | Workpiece defect detection method based on machine vision | |
| CN106846352B (en) | Knife edge picture acquisition method and device for lens analysis force test | |
| CN107516098A (en) | A Method of Extracting 3D Information of Object Outline Based on Edge Curvature Angle | |
| CN105160652A (en) | Handset casing testing apparatus and method based on computer vision | |
| CN109060836A (en) | High-pressure oil pipe joint external screw thread detection method based on machine vision | |
| CN105139384B (en) | The method and apparatus of defect capsule detection | |
| CN114627080B (en) | Vehicle stamping accessory defect detection method based on computer vision | |
| US10074551B2 (en) | Position detection apparatus, position detection method, information processing program, and storage medium | |
| CN116740061B (en) | Visual detection method for production quality of explosive beads | |
| Flesia et al. | Sub-pixel straight lines detection for measuring through machine vision | |
| CN105405122A (en) | Circle detection method based on data stationarity | |
| CN112767497A (en) | High-robustness calibration device based on circular calibration plate and positioning method | |
| CN107230212B (en) | Vision-based mobile phone size measuring method and system | |
| CN104050660A (en) | Method for measuring workpiece round edges | |
| CN106845535B (en) | Typical Components recognition methods based on cloud | |
| CN115880303A (en) | Sub-pixel precision positioning detection method and system for PCB circular hole | |
| CN106556606B (en) | On-line quality control method of large mechanical product assembly based on machine vision | |
| CN111915670A (en) | Vacuum cup size measuring method based on vision | |
| CN105718929B (en) | High-precision and rapid circular target positioning method and system in all-weather unknown environment | |
| CN107818583A (en) | Cross searching detection method and device |
| Date | Code | Title | Description |
|---|---|---|---|
| MM4A | Annulment or lapse of patent due to non-payment of fees |