TWI661394B - Image recognion method for soldered dot - Google Patents

Image recognion method for soldered dot Download PDF

Info

Publication number
TWI661394B
TWI661394B TW107129342A TW107129342A TWI661394B TW I661394 B TWI661394 B TW I661394B TW 107129342 A TW107129342 A TW 107129342A TW 107129342 A TW107129342 A TW 107129342A TW I661394 B TWI661394 B TW I661394B
Authority
TW
Taiwan
Prior art keywords
component
solder joint
distance
components
skirt
Prior art date
Application number
TW107129342A
Other languages
Chinese (zh)
Other versions
TW202009874A (en
Inventor
林佑學
吳鴻威
蔡家銓
詹順凱
Original Assignee
林佑學
吳鴻威
蔡家銓
詹順凱
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 林佑學, 吳鴻威, 蔡家銓, 詹順凱 filed Critical 林佑學
Priority to TW107129342A priority Critical patent/TWI661394B/en
Application granted granted Critical
Publication of TWI661394B publication Critical patent/TWI661394B/en
Publication of TW202009874A publication Critical patent/TW202009874A/en

Links

Landscapes

  • Image Analysis (AREA)

Abstract

一種焊點影像辨識方法,包含:讀取一焊點位置資料,將數焊點投影至一第一平面;讀取一元件資料,將數元件投影至一第二平面;將該第一平面及該第二平面堆疊;計算其中一焊點投影位置與其中一元件投影位置之間的距離,得到一第一距離值;及若該第一距離值超出一第一預設距離,則判斷該焊點屬於不同元件之焊點,若該第一距離值未超出該第一預設距離,則判斷該焊點屬於同一元件之焊點。A solder joint image recognition method includes: reading a solder joint position data and projecting a number of solder joints onto a first plane; reading a component data and projecting a digital element onto a second plane; the first plane and The second plane is stacked; calculating a distance between a projection position of a solder joint and a projection position of a component to obtain a first distance value; and if the first distance value exceeds a first preset distance, judging the weld The points belong to solder joints of different components. If the first distance value does not exceed the first preset distance, it is determined that the solder joints belong to solder joints of the same component.

Description

焊點影像辨識方法Image recognition method of solder joints

本創作係關於一種影像辨識方法,特別是一種焊點影像辨識方法。 This creation is about an image recognition method, especially a solder joint image recognition method.

科技發展迅速,生活中許多便利的設備都離不開科技的輔助,但是現在仍有許多電路的檢測都還是以人工的目視法來檢查,使得產品良率與人力成本不一定成正比。 Technology is developing rapidly, and many convenient devices in life are inseparable from the assistance of technology. However, many circuits are still tested by manual visual inspection, so that the product yield is not necessarily proportional to the labor cost.

焊接不僅起到電氣連接的作用,同時也是固定元件的手段。但是,隨著元件的微型化,焊點間的距離也越來越近。為了維持元件的品質,焊點的檢查,變成了一個重要的課題。 Welding not only serves as an electrical connection, it is also a means of fixing components. However, as components are miniaturized, the distance between solder joints is getting closer. In order to maintain component quality, solder joint inspection has become an important issue.

本創作之一目的係提供一種焊點影像辨識方法,能夠快速地判斷焊點與元件之歸屬。 One of the purposes of this creation is to provide a method for identifying solder joint images, which can quickly determine the ownership of solder joints and components.

本創作之另一目的係提供一種焊點影像辨識方法,藉由雙重投影,能夠快速地判斷焊點與元件之歸屬。 Another purpose of this creation is to provide a solder joint image identification method, which can quickly determine the assignment of solder joints and components through dual projection.

本創作之另一目的係提供一種焊點影像辨識方法,藉由裙邊投影,能夠快速地且更準確地,進一步判斷焊點與元件之歸屬。 Another purpose of this creation is to provide a solder joint image identification method, which can quickly and more accurately judge the ownership of solder joints and components by using skirt projection.

本創作之另一目的係提供一種焊點影像辨識方法,藉由雙重投影、裙邊投影,能夠減少焊點與元件之歸屬誤判。 Another purpose of this creation is to provide a solder joint image identification method, which can reduce the misjudgment of solder joints and components through dual projection and skirt projection.

為達到前述及其他目的,本創作提供之技術手段,包含:一種焊點影像辨識方法,包含:讀取一焊點位置資料,將數焊點投影至一第一平面;讀取一元件資料,將數元件投影至一第二平面;將該第一平面及該第二平面堆疊;計算其中一焊點投影位置與其中一元件投影位置之間的距離,得到一第一距離值;及若該第一距離值超出一第一預設距離,則判斷該焊點屬於不同元件之焊點,若該第一距離值未超出該第一預設距離,則判斷該焊點屬於同一元件之焊點。 In order to achieve the foregoing and other objectives, the technical means provided by this creation include: a method for identifying a solder joint image, including: reading a solder joint position data, projecting a few solder joints onto a first plane; reading a component data, Projecting a digital element onto a second plane; stacking the first plane and the second plane; calculating a distance between a projection position of a solder joint and a projection position of one element to obtain a first distance value; and if the If the first distance value exceeds a first preset distance, it is judged that the solder joint belongs to a solder joint of a different component. If the first distance value does not exceed the first preset distance, it is judged that the solder joint belongs to a solder joint of the same component. .

在一實施例中,所述之焊點影像辨識方法,其中讀取元件資料步驟包含從資料庫中讀取元件型號、元件尺寸或元件預設布局位置座標。 In an embodiment, the solder joint image recognition method, wherein the step of reading component data comprises reading a component model, a component size, or a preset layout position coordinate of the component from a database.

在一實施例中,所述之焊點影像辨識方法,其中讀取元件資料步驟包含從資料庫中讀取相對應機台之安全作業距離。 In one embodiment, the solder joint image recognition method, wherein the step of reading component data includes reading a safe working distance of a corresponding machine from a database.

在一實施例中,所述之焊點影像辨識方法,另包含:從該第二平面中,尋找各元件之中心座標;計算各元件中心座標之距離;及若各元件中心座標之距離超出一第二預設距離,則判斷該些元件屬於一安全群組,若各元件中心座標之距離未超出該第二預設距離,則判斷該些元件屬於一不安全群組。 In one embodiment, the method for identifying a solder joint image further includes: finding the center coordinates of each component from the second plane; calculating the distance between the center coordinates of each component; and if the distance between the center coordinates of each component exceeds one For the second preset distance, it is determined that the components belong to a safe group. If the distance between the center coordinates of the components does not exceed the second preset distance, the components are determined to belong to an unsafe group.

在一實施例中,所述之焊點影像辨識方法,另包含:從該不安全群組中,選擇其中一元件;從該焊點位置資料及該元件資料中,讀取一高度表;以該元件之投影及該高度表,生成該元件之裙邊投影;將各元件之裙邊投影疊 加;及若各元件之裙邊投影不重疊時,則判斷該些元件屬於一次安全群組,若各元件之裙邊投影重疊時,則判斷該些元件屬於一高度不安全群組。 In one embodiment, the method for identifying a solder joint image further includes: selecting one of the components from the unsafe group; reading an altimeter from the solder joint position data and the component data; and The projection of the component and the altimeter generate the skirt projection of the component; the projection of the skirt of each component is superimposed Plus; and if the projections of the skirts of the components do not overlap, the components are judged to belong to a security group; if the projections of the skirts of the components are overlapped, the components are judged to belong to a highly insecure group.

在一實施例中,所述之焊點影像辨識方法,其中生成該元件之裙邊投影之步驟包含以該高度表所設定之安全距離之1/2距離,生成該元件之裙邊投影。 In one embodiment, the solder joint image recognition method, wherein the step of generating a skirt projection of the component includes generating a skirt projection of the component at a distance of 1/2 of the safety distance set by the altimeter.

在一實施例中,所述之焊點影像辨識方法,另包含:將每一元件之裙邊投影編號,並各賦予一裙邊參數,該裙邊參數為(X,Y,L),其中X,Y為座標值;及將每一元件之裙邊投影,各賦予一亮度值,該亮度值為該裙邊參數之L,該亮度值預設為1。 In an embodiment, the method for identifying a solder joint image further includes: projecting a skirt number of each component and assigning a skirt parameter to each skirt, the skirt parameter is (X, Y, L), where X and Y are coordinate values; and each component's skirt is projected, each assigned a brightness value, the brightness value is L of the skirt parameter, and the brightness value is preset to 1.

在一實施例中,所述之焊點影像辨識方法,其中該將各元件之裙邊投影疊加步驟包含將各元件疊加後之裙邊參數之亮度值相加,若相加後亮度值大於2,則判斷各元件之裙邊投影重疊。 In one embodiment, the solder joint image identification method, wherein the step of projecting and superimposing the skirt of each component includes adding the brightness values of the skirt parameters after the components are superimposed. , It is determined that the skirt projections of the components overlap.

在一實施例中,所述之焊點影像辨識方法,另包含:針對該不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。 In one embodiment, the solder joint image recognition method further includes: for the components of the unsafe group, reading the component data from the database again to calculate the actual distance of the components.

在一實施例中,所述之焊點影像辨識方法,另包含:針對該高度不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。 In one embodiment, the solder joint image recognition method further includes: for the components of the highly unsafe group, reading the component data from the database again to calculate the actual distance of the components.

在一實施例中,所述之焊點影像辨識方法,另包含:針對該次安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。 In one embodiment, the solder joint image recognition method further includes: for the components of the security group, reading the component data from the database again to calculate the actual distance of the components.

S1‧‧‧焊點位置投影步驟 S1‧‧‧Soldering Spot Projection Procedure

S2‧‧‧元件位置投影步驟 S2‧‧‧Component location projection steps

S3‧‧‧投影堆疊步驟 S3‧‧‧Projection stacking steps

S4‧‧‧第一距離計算步驟 S4‧‧‧First distance calculation steps

S5‧‧‧第一歸組步驟 S5‧‧‧The first grouping step

S6‧‧‧中心座標步驟 S6‧‧‧ center coordinate steps

S7‧‧‧第二距離計算步驟 S7‧‧‧Second distance calculation steps

S8‧‧‧大部篩選步驟 S8‧‧‧ Most screening steps

S9‧‧‧小部篩選步驟 S9‧‧‧Small selection steps

S10‧‧‧高度表讀取步驟 S10‧‧‧ Altimeter Reading Procedure

S11‧‧‧裙邊投影步驟 S11‧‧‧skirt projection steps

S12‧‧‧裙邊投影疊加步驟 S12‧‧‧skirt projection superposition steps

S13‧‧‧第二歸組步驟 S13‧‧‧Second grouping steps

S141‧‧‧細部篩選步驟 S141‧‧‧Detailed screening steps

S142‧‧‧細部篩選步驟 S142‧‧‧Detailed screening steps

S143‧‧‧細部篩選步驟 S143‧‧‧Detailed screening steps

B‧‧‧焊點 B‧‧‧Solder

C‧‧‧焊點 C‧‧‧Solder

D‧‧‧元件 D‧‧‧Element

E‧‧‧元件 E‧‧‧Element

D1、D2、D3‧‧‧距離 D1, D2, D3‧‧‧ distance

D’‧‧‧裙邊投影 D’ ‧‧‧ skirt projection

X‧‧‧裙邊參數 X‧‧‧skirt parameters

Y‧‧‧裙邊參數 Y‧‧‧skirt parameters

L‧‧‧裙邊參數 L‧‧‧skirt parameters

Φ‧‧‧重疊狀況 Φ‧‧‧ Overlap

圖1是本創作實施例之焊點影像辨識方法之步驟流程圖。 FIG. 1 is a flowchart of steps of a method for identifying a solder joint image according to an embodiment of the present invention.

圖1-1是本創作實施例之投影堆疊步驟結果示意圖。 Figure 1-1 is a schematic diagram of the results of the projection stacking steps of this creative embodiment.

圖2是本創作另一實施例之焊點影像辨識方法之步驟流程圖。 FIG. 2 is a flowchart of steps of a method for identifying a solder joint image according to another embodiment of the present invention.

圖2-1是本創作另一實施例之大部篩選步驟結果示意圖。 Figure 2-1 is a schematic diagram showing the results of most screening steps in another embodiment of the present invention.

圖3是本創作另一實施例之焊點影像辨識方法之步驟流程圖。 FIG. 3 is a flowchart of steps of a method for identifying a solder joint image according to another embodiment of the present invention.

圖3-1是本創作另一實施例之裙邊投影步驟示意圖。 Figure 3-1 is a schematic diagram of a skirt projection step according to another embodiment of the present invention.

圖3-2是本創作另一實施例之裙邊投影步驟示意圖。 FIG. 3-2 is a schematic diagram of a skirt projection step according to another embodiment of the present invention.

本實施方式,係特別針對PCB設計舉例。在PCB設中,Gerber是一種二維向量圖檔格式,它是印刷線路板行業軟體中用於描述印刷線路板圖像的標準格式。Gerber格式的應用之一是將PCB的設計資料從設計轉換到生產。PCB通常由線路板設計人員使用專業的電子設計自動化(EDA)或者電腦輔助設計(CAD)軟體進行設計,然後輸出成Gerber格式檔案。Gerber檔案被送到PCB工廠,並被輸入到電腦輔助製造CAM系統對設計資訊進行轉換,為PCB生產的每一道工藝流程提供資料。 This embodiment is specifically directed to PCB design examples. In PCB design, Gerber is a two-dimensional vector image file format, which is a standard format used to describe printed circuit board images in the software of the printed circuit board industry. One of the applications of the Gerber format is to transform PCB design materials from design to production. PCB is usually designed by circuit board designers using professional electronic design automation (EDA) or computer-aided design (CAD) software, and then exported to Gerber format files. The Gerber file is sent to the PCB factory and is input to the computer-aided manufacturing CAM system to convert the design information and provide data for each process of PCB production.

另,下文所述「投影」,本領域具有通常知識者,可以藉由採用「CAM350」軟體來輔助進行,其細節下文不予贅述。 In addition, the "projection" described below can be carried out by those with ordinary knowledge in the field by using "CAM350" software. The details are not described below.

圖1是本創作實施例之焊點影像辨識方法之步驟流程圖。本實施例之焊點影像辨識方法,包含:焊點位置投影步驟S1,讀取一焊點位置資料,將數焊點投影至一第一平面;元件位置投影步驟S2,讀取一元件資料,將數元件投影至一第二平面;投影堆疊步驟S3,將該第一平面及該第二平面堆疊;第一距離計算步驟S4,計算其中一焊點投影位置與其中一元件投影位置之間的距離,得到一第一距離值;及第一歸組步驟S5,若該第一距離值超出一第一預設 距離,則判斷該焊點屬於不同元件之焊點,若該第一距離值未超出該第一預設距離,則判斷該焊點屬於同一元件之焊點。 FIG. 1 is a flowchart of steps of a method for identifying a solder joint image according to an embodiment of the present invention. The method for identifying a solder joint image in this embodiment includes: a solder joint position projection step S1, reading a solder joint position data, and projecting several solder joints onto a first plane; a component position projection step S2, reading a component data, Projecting a digital element onto a second plane; projecting and stacking step S3, stacking the first plane and the second plane; first distance calculating step S4, calculating a distance between a projection position of a solder joint and a projection position of one of the elements Distance to obtain a first distance value; and a first grouping step S5, if the first distance value exceeds a first preset Distance, it is judged that the solder joint belongs to a solder joint of a different component, and if the first distance value does not exceed the first preset distance, it is judged that the solder joint belongs to a solder joint of the same component.

圖1-1是本創作實施例之投影堆疊步驟結果示意圖。舉例來說,距離計算步驟S4後,再進行第一歸組步驟S5。若第一距離值D1未超出該第一預設距離,則判斷該焊點屬於同一元件之焊點。再舉例來說,第一距離值D2超出一第一預設距離,則判斷該焊點屬於不同元件之焊點。 Figure 1-1 is a schematic diagram of the results of the projection stacking steps of this creative embodiment. For example, after the distance calculation step S4, the first grouping step S5 is performed. If the first distance value D1 does not exceed the first preset distance, it is determined that the solder joint belongs to a solder joint of the same component. For another example, if the first distance value D2 exceeds a first preset distance, it is determined that the solder joint belongs to a solder joint of a different component.

特別藉由第一歸組步驟S5,可以快速的將元件歸組,增加判斷的效率。 In particular, through the first grouping step S5, the components can be quickly grouped and the efficiency of judgment can be increased.

實施例所述之焊點影像辨識方法,其中讀取元件資料步驟S2包含從資料庫中讀取元件型號、元件尺寸或元件預設布局位置座標。 The solder joint image recognition method according to the embodiment, wherein step S2 of reading component data includes reading a component model, a component size, or a preset layout position coordinate of a component from a database.

實施例所述之焊點影像辨識方法,其中讀取元件資料步驟S2包含從資料庫中讀取相對應機台之安全作業距離。 The method for identifying a solder joint image according to the embodiment, wherein step S2 of reading component data includes reading a safe working distance of a corresponding machine from a database.

圖2是本創作另一實施例之焊點影像辨識方法之步驟流程圖。在本實施例中,所述之焊點影像辨識方法,另包含:中心座標步驟S6,從該第二平面中,尋找各元件之中心座標;第二距離計算步驟S7,計算各元件中心座標之距離;及大部篩選步驟S8,若各元件中心座標之距離超出一第二預設距離,則判斷該些元件屬於一安全群組,若各元件中心座標之距離未超出該第二預設距離,則判斷該些元件屬於一不安全群組。 FIG. 2 is a flowchart of steps of a method for identifying a solder joint image according to another embodiment of the present invention. In this embodiment, the solder joint image recognition method further includes: a center coordinate step S6, searching for the center coordinates of each component from the second plane; a second distance calculating step S7, calculating the center coordinates of each component Distance; and most of the screening steps S8, if the distance between the center coordinates of each component exceeds a second preset distance, it is determined that the components belong to a security group, and if the distance between the center coordinates of each component does not exceed the second preset distance , It is determined that the components belong to an insecure group.

圖2-1是本創作另一實施例之大部篩選步驟結果示意圖。舉例來說,第二距離計算步驟S7後,兩元件中心座標之距離為D3。此時可經過大部篩選步驟S8判斷,若距離D3未超出一第二預設距離,則判斷該些元件屬於一不安全群組。 Figure 2-1 is a schematic diagram showing the results of most screening steps in another embodiment of the present invention. For example, after the second distance calculation step S7, the distance between the center coordinates of the two elements is D3. At this time, it can be determined through most screening steps S8. If the distance D3 does not exceed a second preset distance, it is determined that the components belong to an unsafe group.

特別藉由大部篩選步驟S8,可以快速的將元件歸群,增加判斷的效率。 In particular, through most of the screening steps S8, the components can be quickly grouped and the efficiency of judgment can be increased.

實施例所述之焊點影像辨識方法,另包含:細部篩選步驟S141,針對該不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。藉此,可以增加判斷的精確度。 The solder joint image recognition method described in the embodiment further includes: a detailed screening step S141, and for the components of the unsafe group, the component data is read again from the database to calculate the actual distance of the components. Thereby, the accuracy of judgment can be increased.

圖3是本創作另一實施例之焊點影像辨識方法之步驟流程圖。在本實施例中,所述之焊點影像辨識方法,另包含:小部篩選步驟S9,從該不安全群組中,選擇其中一元件;高度表讀取步驟S10,從該焊點位置資料及該元件資料中,讀取一高度表;裙邊投影步驟S11,以該元件之投影及該高度表,生成該元件之裙邊投影;裙邊投影疊加步驟S12,將各元件之裙邊投影疊加;及第二歸組步驟S13,若各元件之裙邊投影不重疊時,則判斷該些元件屬於一次安全群組,若各元件之裙邊投影重疊時,則判斷該些元件屬於一高度不安全群組。 FIG. 3 is a flowchart of steps of a method for identifying a solder joint image according to another embodiment of the present invention. In this embodiment, the solder joint image recognition method further includes: a small part screening step S9, selecting one of the components from the unsafe group; an altimeter reading step S10, from the solder joint position data And the component data, read an altimeter; skirt projection step S11, using the component's projection and the altimeter to generate the skirt projection of the component; skirt projection superimposing step S12, project the skirt of each component Superimposition; and the second grouping step S13, if the skirt projections of the components do not overlap, it is determined that the components belong to a security group; if the skirt projections of the components are overlapped, the components are determined to belong to a height Insecure group.

圖3-1是本創作另一實施例之裙邊投影步驟示意圖。繼承上述圖2-1的例子,此時進行小部篩選步驟S9,從該不安全群組中,選擇其中一元件。接著舉例來說,根據高度表讀取步驟S10的高度表,元件D及焊點B、C,可經過裙邊投影步驟S11,生成該元件之裙邊投影D’。 Figure 3-1 is a schematic diagram of a skirt projection step according to another embodiment of the present invention. Inheriting the example of FIG. 2-1 described above, a small part screening step S9 is performed at this time, and one of the components is selected from the unsafe group. Then, for example, according to the altimeter reading step S10, the component D and the solder joints B and C can be subjected to the skirt projection step S11 to generate the skirt projection D 'of the component.

實施例所述之焊點影像辨識方法,其中生成該元件之裙邊投影之步驟包含以該高度表所設定之安全距離之1/2距離,生成該元件之裙邊投影。 The method for identifying a solder joint image according to the embodiment, wherein the step of generating a skirt projection of the component includes generating a skirt projection of the component at a distance of 1/2 of the safety distance set by the altimeter.

實施例所述之焊點影像辨識方法,另包含:將每一元件之裙邊投影編號,並各賦予一裙邊參數,該裙邊參數為(X,Y,L),其中X,Y為座標值;及 將每一元件之裙邊投影,各賦予一亮度值,該亮度值為該裙邊參數之L,該亮度值預設為1。 The method for identifying a solder joint image according to the embodiment further includes: projecting a skirt number of each element and assigning a skirt parameter to each skirt, the skirt parameter is (X, Y, L), where X, Y is Coordinate values; and The skirt of each element is projected to give a brightness value, the brightness value is L of the skirt parameter, and the brightness value is preset to 1.

實施例所述之焊點影像辨識方法,其中該將各元件之裙邊投影疊加步驟包含將各元件疊加後之裙邊參數之亮度值相加,若相加後亮度值大於2,則判斷各元件之裙邊投影重疊。 The solder joint image identification method according to the embodiment, wherein the step of projecting and superimposing the skirt of each component includes adding the brightness values of the skirt parameters after superimposing the components, and if the brightness values after the addition are greater than 2, determining each The component's skirt projection overlaps.

圖3-2是本創作另一實施例之裙邊投影步驟示意圖。舉例來說,例如D元件具有裙邊參數(X1,Y1,L1),E元件具有裙邊參數(X2,Y2,L2)。經由裙邊投影疊加步驟S12後,若L1+L2大於2,則可判斷有重疊狀況Φ發生。換句話說,此時D、E兩元件的設定位置距離過近,將會影響焊接。因此,根據第二歸組步驟S13,應該則判斷D、E兩元件屬於一高度不安全群組。 FIG. 3-2 is a schematic diagram of a skirt projection step according to another embodiment of the present invention. For example, for example, the D element has a skirt parameter (X1, Y1, L1), and the E element has a skirt parameter (X2, Y2, L2). After the step S12 of projecting and superimposing the skirt, if L1 + L2 is greater than 2, it can be determined that an overlap condition Φ occurs. In other words, at this time, the set positions of the two components D and E are too close, which will affect the welding. Therefore, according to the second grouping step S13, it should be judged that the two components D and E belong to a highly unsafe group.

之後可再根據細部篩選步驟S142(後述),從資料庫中再次讀取D、E兩元件的資料,計算D、E兩元件的實際距離。或者,因為經過上述步驟之後,基本已經將上千筆錯誤訊息(判斷距離過近,但實際上並非如此)減少為數筆,此時將可以依靠人工作業,來進行最精確的判斷。 After that, according to the detailed screening step S142 (described later), the data of the two components D and E are read again from the database to calculate the actual distance between the two components D and E. Or, after going through the above steps, thousands of error messages (judging distance is too close, but actually not so) have been reduced to several. At this time, manual operations can be used to make the most accurate judgment.

實施例所述之焊點影像辨識方法,另包含:細部篩選步驟S142,針對該高度不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。藉此,可以增加判斷的精確度。 The solder joint image recognition method described in the embodiment further includes: a detailed screening step S142, and for the components of the highly unsafe group, the component data is read again from the database to calculate the actual distance of the components. Thereby, the accuracy of judgment can be increased.

實施例所述之焊點影像辨識方法,另包含:細部篩選步驟S143,針對該次安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。藉此,可以增加判斷的精確度。 The solder joint image recognition method described in the embodiment further includes: a detailed screening step S143, and for the components of the security group, the component data is read again from the database to calculate the actual distance of the components. Thereby, the accuracy of judgment can be increased.

值得一提的是,細部篩選步驟S142、細部篩選步驟S143,可以分開執行,或擇一執行。在擇一執行的情形下,可以僅挑選細部篩選步驟S142執 行。換句話說,即僅針對該高度不安全群組之元件,再次判斷即可。藉此,可以加速判斷的效率。 It is worth mentioning that the detail selection step S142 and the detail selection step S143 may be performed separately or alternatively. In the case of alternative execution, only the detailed filtering step S142 may be performed. Row. In other words, only the components of the highly insecure group can be judged again. Thereby, the efficiency of judgment can be accelerated.

雖然本創作以利用上述實施例說明實施方式,但上述說明並非用以限定本創作,在本領域具有通常知識者應知,再不脫離本創作的精神和範圍之內,依據上述實施方式進行各種更動與修改仍屬本創作所保護之範疇。 Although this creation uses the above-mentioned embodiments to describe the implementation, the above description is not intended to limit the creation, and those with ordinary knowledge in the field should know that without departing from the spirit and scope of this creation, make various changes in accordance with the above-mentioned embodiments Modifications and modifications remain within the scope of this creation.

Claims (11)

一種焊點影像辨識方法,包含: 讀取一焊點位置資料,將數焊點投影至一第一平面; 讀取一元件資料,將數元件投影至一第二平面; 將該第一平面及該第二平面堆疊; 計算其中一焊點投影位置與其中一元件投影位置之間的距離,得到一第一距離值;及 若該第一距離值超出一第一預設距離,則判斷該焊點屬於不同元件之焊點, 若該第一距離值未超出該第一預設距離,則判斷該焊點屬於同一元件之焊點。A solder joint image recognition method includes: reading a solder joint position data, and projecting a solder joint onto a first plane; reading a component data, projecting a solder joint onto a second plane; The second plane stack; calculating a distance between a projection position of a solder joint and a projection position of a component to obtain a first distance value; and determining the solder if the first distance value exceeds a first preset distance The points belong to solder joints of different components. If the first distance value does not exceed the first preset distance, it is determined that the solder joints belong to solder joints of the same component. 依請求項1所述之焊點影像辨識方法,其中讀取元件資料步驟包含從資料庫中讀取元件型號、元件尺寸或元件預設布局位置座標。The solder joint image identification method according to claim 1, wherein the step of reading component data includes reading a component model, a component size, or a preset layout position coordinate of the component from a database. 依請求項1所述之焊點影像辨識方法,其中讀取元件資料步驟包含從資料庫中讀取相對應機台之安全作業距離。The method for identifying a solder joint image according to claim 1, wherein the step of reading component data includes reading a safe working distance of a corresponding machine from a database. 依請求項1所述之焊點影像辨識方法,另包含: 從該第二平面中,尋找各元件之中心座標; 計算各元件中心座標之距離;及 若各元件中心座標之距離超出一第二預設距離,則判斷該些元件屬於一安全群組, 若各元件中心座標之距離未超出該第二預設距離,則判斷該些元件屬於一不安全群組。The solder joint image identification method according to claim 1, further comprising: finding the center coordinates of each component from the second plane; calculating a distance between the center coordinates of each component; and if the distance between the center coordinates of each component exceeds a second If the distance is preset, the components are determined to belong to a safe group. If the distance between the center coordinates of the components does not exceed the second preset distance, the components are determined to belong to an unsafe group. 依請求項4所述之焊點影像辨識方法,另包含: 從該不安全群組中,選擇其中一元件; 從該焊點位置資料及該元件資料中,讀取一高度表; 以該元件之投影及該高度表,生成該元件之裙邊投影; 將各元件之裙邊投影疊加;及 若各元件之裙邊投影不重疊時,則判斷該些元件屬於一次安全群組, 若各元件之裙邊投影重疊時,則判斷該些元件屬於一高度不安全群組。The solder joint image identification method according to claim 4, further comprising: selecting one of the components from the unsafe group; reading an altimeter from the solder joint position data and the component data; using the component The projection of the component and the altimeter to generate the skirt projection of the component; superimpose the projection of the skirt of each component; and if the projection of the skirt of each component does not overlap, determine that the components belong to a security group, if each component When the skirt projections overlap, it is determined that the components belong to a highly insecure group. 依請求項5所述之焊點影像辨識方法,其中生成該元件之裙邊投影之步驟包含以該高度表所設定之安全距離之1/2距離,生成該元件之裙邊投影。The method for identifying a solder joint image according to claim 5, wherein the step of generating a skirt projection of the component includes generating a skirt projection of the component at a distance of 1/2 of the safety distance set by the altimeter. 依請求項5所述之焊點影像辨識方法,另包含: 將每一元件之裙邊投影編號,並各賦予一裙邊參數,該裙邊參數為(X, Y, L),其中X, Y為座標值; 將每一元件之裙邊投影,各賦予一亮度值,該亮度值為該裙邊參數之L,該亮度值預設為1。The solder joint image identification method according to claim 5, further comprising: projecting a skirt number of each component and assigning a skirt parameter to each of the skirt parameters (X, Y, L), where X, Y is the coordinate value; the skirt of each element is projected, and a brightness value is assigned to each, the brightness value is L of the skirt parameter, and the brightness value is preset to 1. 依請求項7所述之焊點影像辨識方法,其中該將各元件之裙邊投影疊加步驟包含將各元件疊加後之裙邊參數之亮度值相加,若相加後亮度值大於2,則判斷各元件之裙邊投影重疊。The solder joint image identification method according to claim 7, wherein the step of projecting and superimposing the skirt of each component includes adding the brightness values of the skirt parameters after the components are superimposed. If the brightness value is greater than 2 after adding, Determine whether the projections of the skirts of the components overlap. 依請求項4之所述之焊點影像辨識方法,另包含: 針對該不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。According to the solder joint image identification method described in claim 4, the method further includes: for the components of the unsafe group, reading the component data from the database again to calculate the actual distance of the components. 依請求項5之所述之焊點影像辨識方法,另包含: 針對該高度不安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。According to the solder joint image identification method described in claim 5, the method further includes: for the components of the highly insecure group, reading the component data from the database again to calculate the actual distance of the components. 依請求項5之所述之焊點影像辨識方法,另包含: 針對該次安全群組之元件,從資料庫中再次讀取元件資料,計算元件實際距離。According to the solder joint image identification method described in claim 5, the method further includes: for the components of the security group, reading the component data from the database again to calculate the actual distance of the components.
TW107129342A 2018-08-22 2018-08-22 Image recognion method for soldered dot TWI661394B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW107129342A TWI661394B (en) 2018-08-22 2018-08-22 Image recognion method for soldered dot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW107129342A TWI661394B (en) 2018-08-22 2018-08-22 Image recognion method for soldered dot

Publications (2)

Publication Number Publication Date
TWI661394B true TWI661394B (en) 2019-06-01
TW202009874A TW202009874A (en) 2020-03-01

Family

ID=67764345

Family Applications (1)

Application Number Title Priority Date Filing Date
TW107129342A TWI661394B (en) 2018-08-22 2018-08-22 Image recognion method for soldered dot

Country Status (1)

Country Link
TW (1) TWI661394B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI741894B (en) * 2020-12-01 2021-10-01 英業達股份有限公司 Abnormal solder point analyzing immediately base on dbscan system and method
TWI744999B (en) * 2020-07-23 2021-11-01 和碩聯合科技股份有限公司 Method for training solder joint detecting model, method for detecting solder joint, and solder joint detecting device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW566255U (en) * 2003-02-11 2003-12-11 Chung Shan Inst Of Science Automatic spot welding apparatus with auxiliary vision features
CN105303579A (en) * 2015-11-11 2016-02-03 广州视源电子科技股份有限公司 Circuit board detection method and device
CN205879113U (en) * 2016-04-18 2017-01-11 颜育仁 Image measuring instrument is in measurement system of encapsulation bonding wire processing procedure

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW566255U (en) * 2003-02-11 2003-12-11 Chung Shan Inst Of Science Automatic spot welding apparatus with auxiliary vision features
CN105303579A (en) * 2015-11-11 2016-02-03 广州视源电子科技股份有限公司 Circuit board detection method and device
CN205879113U (en) * 2016-04-18 2017-01-11 颜育仁 Image measuring instrument is in measurement system of encapsulation bonding wire processing procedure

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI744999B (en) * 2020-07-23 2021-11-01 和碩聯合科技股份有限公司 Method for training solder joint detecting model, method for detecting solder joint, and solder joint detecting device
TWI741894B (en) * 2020-12-01 2021-10-01 英業達股份有限公司 Abnormal solder point analyzing immediately base on dbscan system and method

Also Published As

Publication number Publication date
TW202009874A (en) 2020-03-01

Similar Documents

Publication Publication Date Title
JP5932698B2 (en) Work data generation and inspection method for PCB inspection apparatus
JP6922168B2 (en) Surface mount line quality control system and its control method
CN103091328B (en) System, the device and method of assisting are operated to the sight check of check result
JP2019532377A (en) Printed circuit board metal mask manufacturing method and printed circuit board metal mask manufacturing system
TWI661394B (en) Image recognion method for soldered dot
US8364437B2 (en) Mark arrangement inspecting method, mask data, and manufacturing method of semiconductor device
WO2012096004A1 (en) Solder-attachment inspection method, solder-attachment inspection device, and pcb-inspection system
CN106815444B (en) Method for automatically measuring distance between PCB (printed circuit board) drawing components
CN102681355B (en) Scanning data correcting device and drawing device
JP2008185514A (en) Substrate visual inspection apparatus
CN104765822A (en) 3D database system of electronic device and application of system
KR20120054689A (en) Inspection method
US9105079B2 (en) Method and system for obtaining optical proximity correction model calibration data
JP5045591B2 (en) Method for creating area setting data for inspection area and board appearance inspection apparatus
JP2013234976A (en) External appearance inspection device and external appearance inspection method
KR100994417B1 (en) Method of generating work data for Automated Optical Inspection of surface mounted PCB
CN100410626C (en) Pattern evaluation method, pattern splicing method and computer readable media
Han et al. A template matching based method for surface-mount rectangular-pin-chip positioning and defect detection
JP2008283050A (en) Method and device for generating data for image processing
JP2013211284A (en) Solder print inspection device
JP2010147322A (en) Method of creating 3d mounting data of component mounting machine
JP5194338B2 (en) Masking Gerber data creation device, mask Gerber data creation method, and mask Gerber data creation program for solder printing mask
JP7204504B2 (en) Object confirmation device
KR102177329B1 (en) Method for sensing of fiducial Mark
WO2024045204A1 (en) Method and apparatus for locating production monitoring point of photomask, and electronic device