CN103846539A - Image recognition method - Google Patents

Image recognition method Download PDF

Info

Publication number
CN103846539A
CN103846539A CN 201210498499 CN201210498499A CN103846539A CN 103846539 A CN103846539 A CN 103846539A CN 201210498499 CN201210498499 CN 201210498499 CN 201210498499 A CN201210498499 A CN 201210498499A CN 103846539 A CN103846539 A CN 103846539A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
area
silver
welding
wires
foil
Prior art date
Application number
CN 201210498499
Other languages
Chinese (zh)
Inventor
王婳懿
乔凤斌
Original Assignee
上海航天设备制造总厂
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

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K11/00Resistance welding; Severing by resistance heating
    • B23K11/10Spot welding; Stitch welding
    • B23K11/11Spot welding
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K11/00Resistance welding; Severing by resistance heating
    • B23K11/36Auxiliary equipment
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection

Abstract

The invention aims at providing an image recognition method. Welding spots for welding silver wires on the surface of silver foil are recognized in acquired images. The acquired images comprise at least one area corresponding to the silver foil and at least one second area corresponding to the silver wires. The method comprises the steps of (a) obtaining grey levels of pixels of the first area and the second area; (b) obtaining characteristic points of the pixels in the first area and the second area; (c) searching the welding spots in the second area according to the grey levels of the pixels in the second area. Compared with the prior art, recognition is performed according to the grey difference between the silver foil and the silver wires, and positions of the wires are searched around particles of the silver foil to determine the welding spots finally. By the aid of the method, welding spot positions can be recognized conveniently and rapidly.

Description

图像识别方法 Image Recognition

[0001] _ [0001] _

技术领域 FIELD

[0002] 本发明涉及图像识别领域,尤其涉及一种视觉图像处理方法,用于实现太阳能电池阵的自动焊接。 [0002] The present invention relates to image recognition, and particularly to a visual image processing method for automatic soldering of the solar cell array.

背景技术 Background technique

[0003] 国内空间用太阳能电池阵的零部件焊接基本都采用人工焊接,采用人工定位方式利用电阻点焊在太阳翼帆板背面银箔上焊接银导线,由人眼瞄准焊点后再控制操作盒移动焊接机构,每个焊点都需要重复粗调、瞄准、微调、试下压再微调这一过程才能准确移动到焊点的正上方。 [0003] Domestic and components of the welding space solar arrays are used basically manual welding, artificial targeting silver wire by resistance spot welding on the back of the solar wing panels silver, by the human eyes after a control operation aiming pads welding cartridge moving means, each solder joint need to repeat coarse, aiming, trimming, and then pressed again in order to accurately fine tune the process moves to directly above the pad.

[0004] 这种方法耗时效率低,完成一块银箔上焊点的焊接大约需要2分钟的时间,极大的影响了生产进度。 [0004] This method is time-consuming inefficient, on completion of a silver soldering pads takes about two minutes, a great impact on the production schedule. 同时由于太阳能电池阵表面导线、元器件数量多,人工长时间焊接容易疲劳,所以焊接质量得不到很好保障。 And because the solar array surface of the wire, the number of components and more artificial long fatigue welding, the welding quality can not be so well protected.

[0005] 因此,业界需要一种用于太阳能电池阵自动焊接系统的图像识别模块,以期可定位快速准确,能高效完成太阳能电池阵表面电子元器件的可靠焊接。 [0005] Accordingly, a need in an image recognition module solar array system for automatic welding, in order to be positioned quickly and accurately, efficiently complete the solar array and reliable weld surface of the electronic components.

发明内容 SUMMARY

[0006] 为了解决太阳能电池阵人工定位焊接效率低下、可靠性低的问题,采取自动定位方式对焊点进行定位。 [0006] In order to solve the solar array is manually positioned low welding efficiency, low reliability problems and take automatic positioning of solder positioning mode. 由于焊点的随机分布性,不可能事先给出准确的位置(坐标)信息,须采用机器视觉对焊点的坐标位置进行识别。 Because of the random distribution of the solder joints, it is not possible given in advance the exact position (coordinates) information of the coordinate position of the solder joints shall be identified using machine vision. 本发明提供了图像识别技术,操作人员不需对焊点反复调试、瞄准,提高了操作的效率和可靠性。 The present invention provides an image recognition technology, without the operator repeated testing of the solder joint, aimed at improving the efficiency and reliability of operation.

[0007] 根据本发明的一个方面,提供了一种图像识别方法,其在经采集的采集图像中识别出将银导线焊接在银箔表面上的焊点,所述采集图像包括至少一个对应于所述银箔的第一区域和至少一个对应于所述银导线的第二区域。 [0007] In accordance with one aspect of the invention, there is provided an image recognition method which recognizes a silver wire bonding pads on the surface of the silver foil by the captured image acquisition, the acquired image comprises at least one corresponding a first region of the foil and at least one second region corresponding to said silver wire. 所述图像识别方法包括如下步骤:(a)获取所述第一区域和所述第二区域中各像素的灰度值;(b)基于所述第一区域中各像素的灰度值,获取所述第一区域的特征点•'及(O基于所述特征点,根据所述第二区域中各像素的灰度值在所述第二区域中找寻所述焊点。 The image recognition method comprising the steps of: (a) obtaining the first region and the second region gradation value of each pixel; (b) the first region based on the gradation value of each pixel, obtaining the feature point of the first region • 'and (O, based on the feature points, based on the second region gradation value of each pixel to find the pads in the second region.

[0008] 优选地,所述步骤(a)还包括:(A)对所述第一区域和所述第二区域的各像素的灰度值进行灰度变换,以增大所述第一区域中各像素的灰度值与所述第二区域中各像素的灰度值之间的差异,并且,步骤(A)包括减小所述第一区域中各像素的灰度值,并且增大所述第二区域中各像素的灰度值。 [0008] Preferably, the step (a) further comprises: (A) the gradation value of each pixel of the first region and the second region gradation conversion, to increase the first region the difference between the gradation value of each pixel in the second region of the gradation value of each pixel, and the step (a) comprises reducing the first region in the gradation value of each pixel, and increasing the gradation value of each pixel in the second region.

[0009] 优选地,所述步骤(b)包括:(bl)根据第一预定灰度值,确定所述第一区域的像素面积,从而确定出所述第一区域的面积区域;且化2)获得所述第一区域的面积区域的质心,作为所述特征点。 [0009] Preferably, said step (b) comprises: (bl) a first predetermined gradation value, determines the pixel area of ​​the first region, to determine the area of ​​the region of the first region; and 2 of ) obtaining the area of ​​the region of the first region of the centroid as the feature point.

[0010] 优选地,所述步骤(bl)包括:(bll)对所述采集图像进行区域分割;(bl2)从各分割区域的起始像素进行区域生长,其中所述区域生长准则为,当像素的灰度值超过预定生长阈值时,则进行区域生长,当像素的灰度值低于预定生长阈值时,则即停止生长;且(bl3)筛选所述生长得到的区域,并对筛选结果进行面积统计,以确定所述第一区域的面积区域。 [0010] Preferably, said step (BL) comprises: (bll) the acquired image divided into areas; (BL2) each divided from the starting area of ​​the pixel region growing, wherein said region growing guidelines, when the gray values ​​of pixels exceeds a predetermined growth threshold, region growing is performed, when the gray value of a pixel is below a predetermined threshold value growth, i.e., the growth stops; and (BL3 is) screening said growth region obtained, and screening results for area statistics, to determine the area of ​​the region of the first region.

[0011] 优选地,所述步骤(C)中,所述基于所述特征点包括以所述特征点的横坐标点或纵坐标点为基准。 [0011] Preferably, the step (C), the feature based on the point or points comprises abscissa ordinate point of the feature point as a reference.

[0012] 优选地,所述步骤(C)中,在以所述横坐标点为基准的情况下,沿纵坐标轴在纵坐标遍历范围内进行纵坐标遍历,比较某一纵坐标周围预定纵坐标范围(Ml)内的纵坐标所在像素与所述某一纵坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一纵坐标所在像素有灰度突变,并且该纵坐标与所述作为基准的横坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的纵坐标点与所述横坐标点确定为所述焊点的坐标点。 [0012] Preferably, the step (C), at a reference point in the case where the abscissa, ordinate along the ordinate axis in the traverse range to traverse the vertical axis, the ordinate around a predetermined longitudinal Comparative the ordinate value of the gradation pixel where the pixel is located in a coordinate range in the ordinate (of Ml); if the gradation difference exceeds a set threshold value, where the pixel represents a gray scale mutation ordinate , and the ordinate and the abscissa are determined as a point where the reference point mutation pixels; calculating the point mutation, wherein, when the point mutation count reaches a predetermined value, then the mutant points corresponding to the abscissa and ordinate point coordinate point is determined as the point of the solder.

[0013] 优选地,所述步骤(C)中,在以所述纵坐标点为基准的情况下,沿横坐标轴在纵坐标遍历范围内进行横坐标遍历,比较某一横坐标周围预定纵坐标范围(M2)内的横坐标所在像素与所述某一横坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一横坐标所在像素有灰度突变,并且该横坐标与所述作为基准的纵坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的横坐标点与所述作为基准的纵坐标点确定为所述焊点的坐标点。 [0013] Preferably, the step (C), in the ordinate of the point to the reference case, the abscissa in the traverse range to traverse along the ordinate axis of abscissas, the abscissa around a predetermined longitudinal Comparative where the gray value pixel abscissa coordinate range (M2) and the abscissa where the certain pixel; tone value if the difference exceeds a set threshold, the abscissa representing the gray scale of a pixel where the mutation , and the abscissa and the ordinate the determined point as a reference point where the mutation of pixels; calculating the point mutation, wherein, when the point mutation count reaches a predetermined value, then the mutant the abscissa of the point corresponding to the determined coordinate point of the solder point as a reference point of the ordinate.

[0014] 本发明的再一方面为一种太阳能电池阵的焊接系统,包括:图像采集模块,其用于采集需要焊接的银导线和银箔的图像;图像识别模块,其接收所述图像采集模块所采集的采集图像,并识别出银导线焊接在银箔表面上的焊点,并且所述图像识别模块如前述图像识别装置;焊接机构,其在所述焊点处对所述银导线和所述银箔进行焊接;及运动控制模块,其根据所述图像识别模块所识别出的焊点,控制所述焊接机构运动到所述焊点所在位置,完成所述焊点的焊接。 [0014] In another aspect of the invention is a solar cell array welding system, comprising: an image acquisition module which acquires an image to be welded and silver for silver wire; image recognition module that receives the image acquisition the captured image acquisition module and identifies silver wire bonding pads on the surface of the silver foil, the image recognition module and the preceding image recognition apparatus; welding means, at which the solder and the silver conductor welding the foil; and a motion control module, based on the image recognition module of the identified weld, said welding means is moved to control the position where the solder, completing the welding joints.

[0015] 与现有技术相比,根据本发明实施例的图像识别方法基于银箔与银导线之间的灰度差异进行识别,在银箔的质点周围找寻引导线所在位置,并最终确定焊点,由此可方便、快速地自动识别焊点位置。 [0015] Compared with the prior art, the image recognition method according to an embodiment of the present invention is based on the difference between the gray silver and the silver wire identification, finding the position where the guide wire around the silver particles and finalized welding point, thereby easily and quickly automatic recognition of spot position.

[0016] 结合附图,根据下文的通过示例说明本发明主旨的描述可清楚本发明的其他方面和优点。 [0016] conjunction with the accompanying drawings, the gist of the present invention DESCRIPTION Other aspects and advantages may be apparent according to the invention by way of example below.

附图说明 BRIEF DESCRIPTION

[0017] 通过阅读参照以下附图所作的对非限制性实施例所作的详细描述,本发明的其它特征、目的和优点将会变得更明显: [0017] By reading the following detailed description of the accompanying drawings of non-limiting embodiments, and other features, objects and advantages of the invention will become more apparent:

图1为根据本发明的图像识别方法的流程图; FIG 1 is a flowchart of an image recognition method of the present invention;

图2示出了用以进行识别的采集图像; FIG 2 shows image capture for performing recognition;

图3为所述采集图像的灰度值的直方图; FIG 3 is a histogram of gray values ​​of the acquired image;

图4为经灰度变换处理后的经采集图像; FIG 4 is acquired by the image after gradation conversion processing;

图5为经识别的图像,其中焊点位置为图中导线上的小黑叉所在的位置; 5 is identified image, wherein the position of spot position in FIG small black cross on conductor located;

图6示出了图像识别方法对于不同种类银箔的识别;及图7为根据本发明的图像识别装置的方块图。 FIG 6 illustrates an image recognition method for identifying different types of silver; and FIG. 7 is a block diagram of an image recognition apparatus according to the invention. [0018] 附图中相同或相似的附图标记代表相同或相似的部件。 [0018] In the drawings the same or similar to the same or like reference numerals refer to the components.

具体实施方式 detailed description

[0019] 参见示出本发明实施例的附图,下文将更详细地描述本发明。 [0019] Referring to the accompanying drawings showing an embodiment of the present invention, the present invention is described in more detail below. 然而,本发明可以以许多不同形式实现,并且不应解释为受在此提出之实施例的限制。 However, the present invention may be embodied in many different forms and should not be construed as limiting, proposed by this embodiment. 相反,提出这些实施例是为了达成充分及完整公开,并且使本技术领域的技术人员完全了解本发明的范围。 Instead, these embodiments are proposed in order to achieve full and complete disclosure, and the person skilled in the art fully understand the scope of the invention. 这些附图中,为清楚起见,可能放大了层及区域的尺寸及相对尺寸。 In these figures, for clarity, it may be exaggerated sizes and relative sizes of layers and regions.

[0020] 现参考图1,详细描述根据本发明的图像识别方法。 [0020] Referring now to Figure 1, detailed description of the image recognition method of the present invention.

[0021] 根据本发明实施例的种图像识别方法,其在经采集的采集图像中识别出将银导线焊接在银箔表面上的焊点,所述采集图像包括至少一个对应于所述银箔的第一区域和至少一个对应于所述银导线的第二区域。 [0021] The kinds of image recognition method of an embodiment of the present invention, which was identified in the captured image acquired in the silver wire bonding pads on the surface of the silver foil, the captured image corresponding to the at least one silver first and second regions corresponding to at least one of said silver wire.

[0022] 如图1所示,根据本发明实施例的图像采集方法中, [0022] As shown in FIG 1, according to the present invention an image acquisition method of Example embodiment,

在步骤S101,获取所述第一区域和所述第二区域中各像素的灰度值。 At step S101, obtaining the first region and the second region gradation value of each pixel.

[0023] 本实施例中,可通过对所述采集图像进行灰度直方图的实施例统计,获取所述第一区域和所述第二区域中各像素的灰度值。 [0023] In this embodiment, the acquired image may be performed by embodiments gray histogram statistics embodiment, obtaining the first region and the second region gradation value of each pixel. 图2示出了所述采集图像,其示出了需要识别的重叠的银箔和银导线。 Figure 2 shows the captured image, which shows the overlap of silver and a silver wire to be identified. 如图2所示,银箔与银导线灰度有差异,银箔表面像素灰度值分布比较均匀,银箔表面像素点灰度值大于银导线灰度值。 As shown, the foil 2 with a silver wire gradation differences, the surface of silver pixel grayscale values ​​more evenly distributed, silver gray values ​​is greater than the surface of the pixel gray scale value silver wire.

[0024]图3是所述采集图像的灰度值的直方图。 [0024] FIG. 3 is a histogram of the gradation value of image acquisition. 为统计整个图片的灰度分布情况,对所述采集图像首先进行灰度的直方图计算,从而得到图3。 Statistical distribution of the gradation of the entire image, the acquired image is first gradation histogram calculation, whereby FIG. 从图3可知,灰度值集中分布在30-40,120-140,145-165这三个区域。 It is seen from FIG. 3, gray value 30-40,120-140,145-165 concentrated in three regions. 通过对背景区域、银导线、银箔的灰度值进行提取,可知背景区域灰度值主要分布为30-40,银导线灰度值主要分布在120-140,银箔的灰度值分布在145-165。 By background region, silver wire, silver foil gradation value extracted, it is found mainly background area gradation value of 30-40, a silver wire 120-140 gradation value mainly in the gray value distribution of silver 145-165. 由此可见,银箔、银导线灰度和背景灰度值存在较大的差异,而银导线和银箔的灰度存在较小的差异。 Thus, the presence of silver, silver wire gray background gray value and a large difference, the gray silver wire and silver present in minor differences.

[0025] 由于根据本发明实施例的图像识别方法是基于银箔与银导线之间的灰度差异进行识别,因此,较佳实施例中,作为灰度值获取的代替或者除此之外,还可对所述采集图像的所述第一区域和第二区域进行灰度的线性变换,最后得到对比度增强的图像,增强了银箔与银导线之间的灰度差异,便于后期的识别。 [0025] Since the image recognition method according to an embodiment of the present invention is identified based on the difference between the gray silver and the silver wire, therefore, instead of the embodiment, as the gradation value acquired or in addition to the preferred embodiment, of the acquired image may first region and the second region gradation linear transformation, finally resulting contrast-enhanced images, enhancing the difference between the gray silver and the silver wires, for easy identification later.

[0026] 本实施例中,对所述第一区域和所述第二区域的各像素的灰度值进行灰度变换,以增大所述第一区域中各像素的灰度值与所述第二区域中各像素的灰度值之间的差异。 [0026] In this embodiment, the gradation value of each pixel of the first region and the second region gradation conversion, to increase the first region of each pixel in the gray value the difference between the second region gradation value of each pixel.

[0027] 本实施例中,通过减小所述第一区域中各像素的灰度值,并且增大所述第二区域中各像素的灰度值,以增大所述第一区域中各像素的灰度值与所述第二区域中各像素的灰度值之间的差异。 [0027] In this embodiment, by reducing the first region in the gradation value of each pixel, and increasing the second region in the gradation value of each pixel to increase the area of ​​each of the first the difference between the pixel value and the second gradation region gradation value of each pixel.

[0028] 下文将详细描述灰度变换的处理。 [0028] The gradation conversion processing will hereinafter be described in detail.

[0029] 图4为经灰度变换处理后的经采集图像。 [0029] FIG 4 is acquired by the gradation conversion by the image processing.

[0030] 设定线性灰度变换函数为,其中fA为线性函数的斜率,fB为线性函数在y轴的截距,Da为输入图像的灰度,Db表示输出图像的灰度,图像预处理的主要作用是增强银箔和银导线之间的灰度差异,采取提高银箔灰度值,降低银导线的灰度值的方法增强两者之前的差异。 [0030] as linear gradation conversion function, which is the slope of the linear function fA, fB is a linear function of the y-axis intercept, the gradation of the input image is Da, Db represents a gradation of an output image, image preprocessing the main role is to enhance the difference between the gray silver and the silver wire, silver foil adopted method for improving gradation value, reducing the silver gray value difference reinforcing wire both before. 设定一个fB值,Dm表示银箔输出图像的灰度,Db2表示银导线输出图像的灰度,联立方程组如下: Setting a value fB, Dm represents the output gray silver image, a silver wire Db2 represents the gradation of the output image, the following simultaneous equations:

Figure CN103846539AD00071

在匕的值域范围选择一个匕值,根据fA、fB值大小,对原采集图像进行灰度的线性变换,一次线性变换完毕,再根据公式进行迭代,进行二次线性变换。 Selecting a value in the range of values ​​dagger dagger, according to fA, fB value of the size of the original image acquired gradation linear transformation, a linear transformation is completed, then in accordance with the iteration equation, secondary linear transformation.

[0031] 较佳实施例中个,可对灰度线性变换完毕的图像提高对比度,设定一个fA > I的值,令fB=0,使得图像整体的灰度增大,图像更加明亮。 [0031] In a preferred embodiment, the linear gradation-converted image to increase contrast and a setting fA> value I, so fB = 0, so that the entire image gradation is increased, the image brighter. 从图4中可以看出,经过图像预处理后的图片,背景图片灰度减小,银箔与银导线显得更加明亮,银箔与银导线之间的灰度差异更加明显。 As can be seen from Figure 4, after the picture image preprocessing, image gray background reduced, silver and silver wire appears brighter, the difference between the gray silver and the silver wire is more apparent.

[0032] 步骤S103中,基于所述第一区域中各像素的灰度值,获取所述第一区域的特征点。 [0032] In step S103, the first region based on the gradation value of each pixel, obtaining the feature point of the first region.

[0033] 本实施例中,根据第一预定灰度值,确定所述第一区域的像素面积,从而确定出所述第一区域的面积区域。 [0033] In this embodiment, according to a first predetermined gradation value, determines the pixel area of ​​the first region, to determine the area of ​​the region of the first region.

[0034] 具体地,首先,对所述采集图像进行区域分割。 [0034] Specifically, first, the image capture area division. 将所述采集图像分割成多个区域之后,再确定所述第一区域的像素面积,从而确定出所述第一区域的面积区域。 After the collected image into a plurality of regions, and then determining the pixel area of ​​the first region, thereby determining that the area of ​​a region of the first region.

[0035] 接着,从各分割区域的起始像素进行区域生长,其中所述区域生长准则为,当像素的灰度值超过预定生长阈值时,则进行区域生长,当像素的灰度值低于预定生长阈值时,则即停止生长。 [0035] Next, a region growing from the start pixel of each divided region, wherein the region growing guidelines, when the gray value of a pixel exceeds a predetermined threshold growth, region growing is performed, when the gradation value of the pixel is below growth predetermined threshold value, then stop growing. 本实施例中,选择生长点像素灰度值为50。 In this embodiment, the pixel gray value of 50 to select the growing point. 换言之,选择区域生长准则为像素的灰度值超过50,进行区域生长,低于50即停止生长。 In other words, selective area growth guidelines pixel gray value exceeds 50, region growing, less than 50 stop growing.

[0036] 最后,筛选所述生长得到的区域,并对筛选结果进行面积统计,以确定所述第一区域的面积区域。 [0036] Finally, the screened region growing obtained, and the results of the screening area statistics, to determine the area of ​​the region of the first region. 本实施例中,因银箔所在区域的面积较大,基本在10000以上,因此确定筛选面积的条件阈值为10000,筛选出银箔所在的面积区域。 Embodiment, because of the larger silver area Area according to the present embodiment, substantially 10,000 or more, the determination of threshold criteria to filter area is 10000, screened area of ​​the region where the foil.

[0037] 由于银导线一般在银箔中心点左右,而质心与银箔中心点基本重合,因此以银箔的质心为基准。 [0037] Since the silver foil wires generally around a center point, the centroid substantially coincides with the center point of the silver foil, silver foil thus centroid as a reference. 即,获得所述第一区域的面积区域的质心,作为所述特征点。 I.e., obtain the area of ​​a region of the first region of the centroid as the feature point.

[0038] 步骤S105中,基于所述特征点,根据所述第二区域中各像素的灰度值在所述第二区域中找寻所述焊点。 [0038] In step S105, based on the feature points, based on the second region gradation value of each pixel to find the pads in the second region. 如前所述,由于银导线一般在银箔中心点左右,而质心与银箔中心点基本重合,因此以银箔的质心为基准让横、纵坐标在质心周围范围内进行遍历,寻找焊点。 As described above, since the silver foil wires generally around a center point, the centroid substantially coincides with the center point of the silver foil, silver foil thus centroid as a reference so that the horizontal and vertical coordinates within the traverse range around the centroid, looking welds .

[0039] 本实施例中,所述基于所述特征点包括以所述特征点的横坐标点或纵坐标点为基准。 [0039] In this embodiment, the feature point based on the feature point comprises the abscissa or ordinate point as a reference point.

[0040] 这样,在以所述横坐标点为基准的情况下,沿纵坐标轴在纵坐标遍历范围内进行纵坐标遍历,比较某一纵坐标周围预定纵坐标范围Ml内的纵坐标所在像素与所述某一纵坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一纵坐标所在像素有灰度突变,并且该纵坐标与所述作为基准的横坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的纵坐标点与所述横坐标点确定为所述焊点的坐标点。 [0040] Thus, along the ordinate axis in the traverse ordinate ordinate traverse range to the abscissa in the case where the reference point, where the surrounding pixel ordinate comparing a predetermined ordinate ordinate range Ml and where the certain pixel ordinate gradation value; if the difference value exceeds the threshold gray value, where the pixel represents a gray scale mutation ordinate, and the ordinate as the reference the abscissa of the point where the pixel determined point mutation; calculates the point mutation, wherein, when the point mutation count reaches a predetermined value, then the point mutation corresponding to the cross point of the ordinate coordinate point is determined as the coordinate point of the solder joints.

[0041] 或者,在以所述纵坐标点为基准的情况下,沿横坐标轴在纵坐标遍历范围内进行横坐标遍历,比较某一横坐标周围预定纵坐标范围NI内的横坐标所在像素与所述某一横坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一横坐标所在像素有灰度突变,并且该横坐标与所述作为基准的纵坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的横坐标点与所述作为基准的纵坐标点确定为所述焊点的坐标点。 [0041] Alternatively, to the ordinate at the point as a reference case, the abscissa in the traverse range to traverse along the ordinate abscissa axis, where the abscissa in the comparison pixel abscissa around a predetermined range ordinate NI the gradation value of a pixel where the abscissa; if the difference value exceeds the threshold gray value, where the abscissa represents the certain pixel gray scale mutations, as the abscissa and the reference ordinate point where the pixel determined point mutation; calculates the point mutation, wherein, when the point mutation count reaches a predetermined value, then the point mutation corresponding to the abscissa of a point ordinate reference point to determine the coordinates of the point of the solder.

[0042] 具体来说,对于横置银箔,固定质心横坐标,让纵坐标从比银箔纵坐标最小值多NI个像素宽度的位置遍历到比银箔纵坐标最大值小NI个像素宽度的位置。 [0042] More specifically, for transverse foil, fixed centroid abscissa, ordinate from a position so that the minimum silver ordinate NI than the width of the pixels traversed smaller than silver ordinate NI maximum pixel width s position. 本实施例中,NI可按需设定。 In this embodiment, NI can be set on demand. 然后,设定一个合适的像素宽度为尺子,规定像素宽度为M1,在纵坐标对银箔进行遍历时,与纵坐标纵向距离Ml范围内的纵坐标点的灰度值与该纵坐标的灰度值进行比较。 Then, a suitable set of pixel width of a ruler, a predetermined pixel width of M1, while the ordinate of the foil is traversed gradation value ordinate of longitudinal distance Ml ordinate point within the range of the vertical coordinates of ash values ​​are compared. 当两者灰度差超过设定阈值时,代表此点有灰度的突变,统计突变点的个数,若突变点超过一定个数,认为此纵坐标为焊点所在位置纵坐标,不超过一定个数,让纵坐标接着遍历银箔,直至得到符合要求的点。 Gradation difference exceeds when both the set threshold, the representative gray scale of this point mutation, count the number of point mutations, point mutations if more than a certain number, that this vertical axis is the ordinate weld location, not more than predetermined number, the ordinate so traversing the foil Next, until a point to meet the requirements. 求得的纵坐标加上质心横坐标即得到银导线上焊点坐标。 Plus obtained ordinate centroid coordinates on the abscissa to obtain the silver solder wire.

[0043] 类似地,竖置银箔上银导线的识别方法与横置银导线相似。 [0043] Similarly, the vertical set of silver foil recognition method is similar to the transverse wire silver wire. 固定质心纵坐标,让横坐标从比银箔横坐标最小值多N2个像素宽度的位置遍历到比银箔横坐标最大值小N2个像素宽度的位置。 Ordinate centroid is fixed, so that from a position of minimum abscissa silver abscissa than the N2 position traversed pixel width smaller than the silver abscissa N2 pixels width maximum. 本实施例中,N2可按需设定。 In this embodiment, N2 may be set on demand. 然后设定一个合适的像素宽度为尺子,规定像素宽度为M2,在横坐标对银箔进行遍历时,与横坐标横向距离M2范围内的横坐标点的灰度值与该横坐标灰度值进行比较,两者之间灰度差超过设定阈值,代表此点有灰度的突变。 And set an appropriate ruler pixel width, a predetermined pixel width of M2, the abscissa of the foil when the traverse, the gradation value in the abscissa point M2 lateral distance range with the abscissa abscissa gradation value comparing the difference between the gradation exceeds the set threshold value, the representative gray scale of this point mutation. 统计突变点的个数,若突变点超过一定个数,代表此横坐标为焊点所在位置横坐标,不超过一定个数,让横坐标接着遍历银箔,直至得到符合要求的点。 The number of point mutations statistics, if more than a certain number of point mutations, on behalf of the abscissa is the location of abscissa joints, not to exceed a certain number, so that the horizontal axis and then traverse the foil until a point to meet the requirements. 求得的横坐标加上质心纵坐标即得到银导线焊点坐标。 Plus obtained abscissa ordinate centroid to obtain a silver solder wire coordinates.

[0044] 较佳实施例中,为提高焊接速率,减少操作复杂度,采取一次操作识别所采集图像上所有银箔焊点坐标。 [0044] In the preferred embodiment, in order to improve the welding rate, reduce operational complexity, all the silver solder joints take a coordinate on the image recognition of the acquired operation. 即,可对包括横置银箔、竖置银箔、采集图像不完整的银箔进行识别。 I.e., it can be recognized foil including transverse, vertical set silver foil, silver foil incomplete image acquisition.

[0045] 图5是对图2图像采集到的图经过灰度变换、图像识别而最终获得的焊点识别图,焊点位置为图中导线上的小黑叉所在的位置,图6展示了图像识别模块对于不同种类银箔的识别。 [0045] FIG. 5 is a solder recognition of the image of FIG. 2 to FIG acquired through gradation conversion, and finally obtained image recognition, the position of spot position in FIG small black cross on conductor located, Figure 6 shows the the image recognition module for recognition of different kinds of silver.

[0046] 现参考图7描述根据本发明的图像识别装置,其在经采集的采集图像中识别出将银导线焊接在银箔表面上的焊点,所述采集图包括至少一个对应于所述银箔的第一区域和至少一个对应于所述银导线的第二区域。 [0046] Referring now to FIG 7 describe the image recognition apparatus according to the invention, which recognizes the acquired image by the acquisition of silver wire bonding pads on the surface of the silver foil, the acquisition includes at least a FIG corresponding to the a first region and a second region of the silver foil corresponding to at least one of said silver wire.

[0047] 如图7所示,所述图像识别装置包括灰度值获取单元,其获取所述第一区域和所述第二区域中各像素的灰度值;特征提取单元,其基于所述第一区域中各像素的灰度值,获取所述第一区域的特征点;及焊点确定单元,其根据所述第二区域中各像素的灰度值在所述第二区域中找寻所述焊点。 [0047] 7, the image recognition apparatus comprises a gradation value acquisition unit that acquires the first region and the second region of each pixel gray value; feature extraction unit, based on the a first region gradation value of each pixel, obtaining the feature point of the first region; and determining means pad, said second region based on the gradation value of each pixel of job in the second region said pad.

[0048] 本实施例中,所述灰度值获取单元对所述第一区域和所述第二区域的各像素的灰度值进行灰度变换,以增强所述第一区域中各像素的灰度值与所述第二区域中各像素的灰度值之间的差异。 [0048] In this embodiment, the gradation value acquisition unit for gradation values ​​of each pixel of the first region and the second region gradation conversion, to enhance the first region of each pixel the difference between the gradation value and the second region gradation value of each pixel. 此外,所述灰度值获取单元减小所述第一区域中各像素的灰度值,并且增大所述第二区域中各像素的灰度值。 Furthermore, the gradation value acquiring unit decreases the first region gradation value of each pixel, and increasing the second region in each pixel gray value.

[0049] 本实施例中,所述特征提取单元根据第一预定灰度值,确定所述第一区域的像素面积,从而确定出所述第一区域的面积区域。 [0049] In this embodiment, the feature extraction unit according to a first predetermined gradation value, determines the pixel area of ​​the first region, thereby determining that the area of ​​a region of the first region. 并且所述特征提取单元获得所述第一区域的面积区域的质心,作为所述特征点。 And the feature extracting unit obtains an area of ​​the first area centroid, as the feature point. [0050] 本实施例中,所述特征提取单元对所述采集图像进行区域分割,并且从各分割区域的起始像素进行区域生长,其中所述区域生长准则为,当像素的灰度值超过预定生长阈值时,则进行区域生长,当像素的灰度值低于预定生长阈值时,则即停止生长。 [0050] In this embodiment, the feature extraction unit for the acquired image segmentation region, and region growing from the start pixel of each divided region, wherein the region growing guidelines, when the gradation value of a pixel exceeds growth predetermined threshold value, the region growing, when the gray value of a pixel is below a predetermined growth threshold, then stop growing. 此外,所述特征提取单元筛选所述生长得到的区域,并对筛选结果进行面积统计,以确定所述第一区域的面积区域。 Furthermore, the feature extraction unit area obtained by growth of the screening, and screening results for area statistics, to determine the area of ​​the region of the first region.

[0051] 本实施例中,所述焊点确定单元的所述基于所述特征点包括以所述特征点的横坐标点或纵坐标点为基准。 [0051] In this embodiment, the pad of the determination unit based on the feature point comprises a point or abscissa ordinate point of the feature point as a reference.

[0052] 本实施例中,所述焊点确定单元中,在以所述横坐标点为基准的情况下,沿纵坐标轴在纵坐标遍历范围内进行纵坐标遍历,比较某一纵坐标周围预定纵坐标范围(Ml)内的纵坐标所在像素与所述某一纵坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一纵坐标所在像素有灰度突变,并且该纵坐标与所述作为基准的横坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的纵坐标点与所述横坐标点确定为所述焊点的坐标点。 [0052] In this embodiment, the solder joint determination unit, to the abscissa in the case where the reference point, along the ordinate axis in the traverse ordinate ordinate traverse range around a comparison ordinate the predetermined range where the pixel ordinate ordinate (of Ml) in the gradation value of a pixel located ordinate; tone value if the difference exceeds a set threshold value, where the pixel represents a gray ordinate degree of mutation, and the ordinate and the abscissa as a reference point where the determined pixel point mutation; calculates the point mutation, wherein, when the point mutation count reaches a predetermined value, then the point mutation corresponding to the abscissa and ordinate point coordinate point is determined as the point of the solder joints.

[0053] 本实施例中,所述焊点确定单元中,在以所述纵坐标点为基准的情况下,沿横坐标轴在纵坐标遍历范围内进行横坐标遍历,比较某一横坐标周围预定纵坐标范围(NI)内的横坐标所在像素与所述某一横坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一横坐标所在像素有灰度突变,并且该横坐标与所述作为基准的纵坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的横坐标点与所述作为基准的纵坐标点确定为所述焊点的坐标点。 [0053] In this embodiment, the solder joint determination unit, to the ordinate at the point as a reference case, the abscissa in the traverse range to traverse along the ordinate axis of abscissas, the abscissa around a comparison gradation values ​​of pixels located within a predetermined abscissa ordinate range (NI) and the abscissa where the certain pixel; gradation value if the difference exceeds a set threshold, the certain pixel location represents the abscissa of gray degree of mutation, and the abscissa the determined pixels located point mutation as ordinate reference point; when calculating the point mutation, wherein the mutation point when the count reaches a predetermined value, then the point mutation corresponding to the abscissa of the point is determined as the coordinate point of the solder as ordinate reference point.

[0054] 根据本发明实施例的图像识别装置的操作与其可实现的功能与根据本发明实施例的图像识别方法相对应,再此不进行过多的描述。 [0054] The functional operation of the image recognition apparatus according to an embodiment of the present invention may be implemented with its corresponding image recognition method according to an embodiment of the present invention, further description will not be excessive.

[0055] 现描述根据本发明实施例的太阳能电池阵的自动焊接系统。 [0055] Automatic welding system will now be described in accordance with an embodiment of the solar array of the present invention. 所述自动焊接系统包括图像采集模块,其用于采集需要焊接的银导线和银箔的图像;图像识别模块,其接收所述图像采集模块所采集的采集图像,并识别出银导线焊接在银箔表面上的焊点;焊接机构,其在所述焊点处对所述银导线和所述银箔进行焊接;及运动控制模块,其根据所述图像识别模块所识别出的焊点,控制所述焊接机构运动到所述焊点所在位置,完成所述焊点的焊接。 The automatic welding system includes an image acquisition module that requires image acquisition soldering silver wire and silver foil is used; an image recognition module, an image capture module that receives said acquired image capture, and identify the wire bonding of silver in silver pads on the surface of the foil; welding means, in which the solder joints of the silver wire and the silver solder; and motion control module, based on the image recognition module identified solder joints, control said welding means moves to a position where the solder, completing the welding joints.

[0056] 本实施例中,所述图像采集模块包括低角度光源,以在采集时利用所述低角度光源对所述银导线和银箔进行照明。 [0056] In this embodiment, the image acquisition module comprises a low angle light source, using said low angle light source is illuminated in the collection of the silver wire and silver foil. 由于银导线与银箔表面颜色都为银白色,差异很小,但银导线在银箔表面,两者高度不一致,为识别银箔表面的银导线,采用低角度光源。 Since the surface of silver wire and silver colors are silver, differences are small, but in the silver foil conductor surface, both the high degree of inconsistency, to identify a silver foil conductor surface, using low angle light. 为使银箔表面灰度均匀,减少明暗差异,便于后期图像识别,采用环形光源,本实施例选择环形低角度光源。 For silver uniform gray surface, reducing shade difference, to facilitate the post image recognition, using the ring light source, the present embodiment selects an annular low angle light.

[0057] 所述图像识别模块如前述权利要求1〜7中任一项所述的图像识别装置,并且所述运动控制模块和焊接机构为业界常见结果,在此不再赘述。 The [0057] image recognition module as claimed in the preceding image recognition apparatus as claimed in any one of claims 1~7, and the motion control module and the welding means is a common result of the industry, are not repeated here.

[0058] 本发明具有如下优点: [0058] The present invention has the following advantages:

(1)根据本发明实施例的图像识别方法基于银箔与银导线之间的灰度差异进行识别,在银箔的质点周围找寻引导线所在位置,并最终确定焊点,由此可方便、快速地自动识别焊点位置; (1) The image recognition method of the present invention is based on the embodiment of the gradation difference between the silver and the silver wire identification, finding the position where the guide wire around the silver particles, and ultimately determine the pad, whereby easy, quick automatic recognition of spot position;

(2)根据本发明实施例的图像识别方法,通过对图像进行灰度直方图的统计,再对图像进行一系列灰度的线性变换,最后得到对比度增强的图像,增强了银箔与银导线之间的灰度差异,便于后期的识别; (2) The image recognition method of the embodiment of the present invention, by the image histogram statistics, and then the image of the linear conversion range of the gradation, contrast-enhanced images finally obtained, the silver wire and silver enhanced the difference between the gray scale, to facilitate later identification;

(3)根据本发明实施例的太阳能电池阵自动焊接系统,能实现对所需的焊点的准确定位,从而实现自动焊接。 (3) The solar cell array of the automatic welding system according to an embodiment of the present invention, to achieve the required accurate positioning of the solder, thereby achieving automatic welding. 采用图像识别模块,操作人员不需对焊点反复调试、瞄准,提高了操作的效率和可靠性; Employing the image recognition module, the operator does not need repeated testing of the solder joint, aimed at improving the efficiency and reliability of operation;

(4)根据本发明实施例的太阳能电池阵自动焊接系统块,由于图像识别模块对光源的要求很高,光源的配置直接影响了后期的图像预处理和图像分析。 (4) The solar cell array block automatic welding system of the present embodiment of the invention, the image recognition module due to the high requirements on the light source, the light source arranged directly affects the image preprocessing and image analysis later. 图像识别模块需要识别银箔表面的银导线,而银箔与银导线颜色相同,普通光源很难识别,银箔和银导线的高度不一致,采用低角度环形光源可增强导线和银箔灰度的对比度。 The image recognition module needs to recognize silver foil conductor surface, whereas the same color of silver with a silver wire, ordinary light is difficult to identify, and the high degree of inconsistency silver silver wire, low angle light source may be an annular reinforcing wire and silver gradation contrast.

[0059] 对于本领域技术人员而言,显然本发明不限于上述示范性实施例的细节,而且在不背离本发明的精神或基本特征的情况下,能够以其他的具体形式实现本发明。 In the case [0059] to those skilled in the art, that the invention is not limited to the details of the above-described exemplary embodiment, but without departing from the spirit or essential characteristics of the present invention, the present invention can be realized in other specific forms. 因此,无论从哪一点来看,均应将实施例看作是示范性的,而且是非限制性的,本发明的范围由所附权利要求而不是上述说明限定,因此旨在将落在权利要求的等同要件的含义和范围内的所有变化囊括在本发明内。 Therefore, no matter from what point of view, the embodiments should be considered exemplary, and not limiting, the scope of the invention being indicated by the appended claims rather than by the foregoing description, the appended claims are therefore intended to All changes which come within the meaning and range of equivalents thereof are within the present invention include. 不应将权利要求中的任何附图标记视为限制所涉及的权利要求。 In the claims should not be considered as any reference numerals as claimed in claim limitations involved. 此夕卜,显然“包括” 一词不排除其他单元或步骤,单数不排除复数。 Bu this evening, apparently "comprising" does not exclude other elements or steps, the singular does not exclude a plurality. 系统权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。 A plurality of units or means recited in the claims the system can also be implemented by a single unit or through software or hardware. 第一,第二等词语用来表示名称,而并不表示任何特定的顺序。 The first, second, etc. are used to indicate the name, but does not indicate any particular sequence.

Claims (9)

  1. 1.一种图像识别方法,其在经采集的采集图像中识别出将银导线焊接在银箔表面上的焊点,所述采集图像包括至少一个对应于所述银箔的第一区域和至少一个对应于所述银导线的第二区域,其特征在于,包括如下步骤: Ca)获取所述第一区域和所述第二区域中各像素的灰度值; (b)基于所述第一区域中各像素的灰度值,获取所述第一区域的特征点;及(c)基于所述特征点,根据所述第二区域中各像素的灰度值在所述第二区域中找寻所述焊点。 An image recognition method which recognizes the acquired image by the acquisition of a silver wire bonding pads on the surface of the silver foil, the first captured image corresponding to the at least one region comprises silver and at least a second region corresponding to said silver wire, characterized by comprising the steps of: Ca) obtaining the first region and the second region gradation value of each pixel; (b) based on the first region gradation value of each pixel, obtaining a characteristic point of the first region; and (c) based on the feature points to find the second region in the second region according to the gradation value of each pixel said welds.
  2. 2.根据权利要求1所述的识别方法,其特征在于,步骤(a)还包括: (A)对所述第一区域和所述第二区域的各像素的灰度值进行灰度变换,以增大所述第一区域中各像素的灰度值与所述第二区域中各像素的灰度值之间的差异, 并且,步骤(A)包括减小所述第一区域中各像素的灰度值,并且增大所述第二区域中各像素的灰度值。 2. The identification method according to claim 1, wherein step (a) further comprises: (A) the gradation value of each pixel of the first region and the second region gradation conversion, to increase the difference between the first region and the gradation value of each pixel in the second region gradation value of each pixel, and the step (a) comprises reducing the first region of each pixel gradation value, and increases said second region in each pixel gray value.
  3. 3.根据权利I所述的识别方法,其特征在于,所述步骤(b)包括: (bl)根据第一预定灰度值,确定所述第一区域的像素面积,从而确定出所述第一区域的面积区域;且(b2)获得所述第一区域的面积区域的质心,作为所述特征点。 3. The identification method according to claim I, wherein said step (b) comprises: (bl) a first predetermined gradation value, determines the pixel area of ​​the first region, the second to determine the area of ​​a region of a region; and (b2) obtaining the area of ​​the region of the first region of the centroid as the feature point.
  4. 4.根据权利3所述的识别方法,其特征在于,所述步骤(bl)包括: (bll)对所述采集图像进行区域分割; (bl2)从各分割区域的起始像素进行区域生长,其中所述区域生长准则为,当像素的灰度值超过预定生长阈值时,则进行区域生长,当像素的灰度值低于预定生长阈值时,则即停止生长;且(bl3)筛选所述生长得到的区域,并对筛选结果进行面积统计,以确定所述第一区域的面积区域。 The identification method of claim 3, wherein said step (BL) comprises: (bll) the acquired image divided into areas; (BL2) for region growing from the start pixel of each divided region, wherein said region growing guidelines, when the gray value of a pixel exceeds a predetermined threshold growth, region growing is performed, when the gray value of a pixel is below a predetermined threshold value growth, i.e., the growth stops; and (BL3 is) screening said the resulting growth area, and the area of ​​the screening results statistically, to determine the area of ​​the region of the first region.
  5. 5.根据权利I所述的识别方法,其特征在于,所述步骤(c)中,所述基于所述特征点包括以所述特征点的横坐标点或纵坐标点为基准。 5. The identification method according to claim I, wherein said step (c), the abscissa based on the feature points comprise points of the feature point or a reference point ordinate.
  6. 6.根据权利5所述的识别方法,其特征在于,所述步骤(c)中,在以所述横坐标点为基准的情况下,沿纵坐标轴在纵坐标遍历范围内进行纵坐标遍历,比较某一纵坐标周围预定纵坐标范围(Ml)内的纵坐标所在像素与所述某一纵坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一纵坐标所在像素有灰度突变,并且该纵坐标与所述作为基准的横坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的纵坐标点与所述横坐标点确定为所述焊点的坐标点。 The identification method according to claim 5, wherein said step (c), in the abscissa to the case where the reference point, along the ordinate axis in the ordinate traverse range to traverse the ordinate , where the gray value around the pixel ordinate the ordinate comparing a predetermined range ordinate (of Ml) with the certain pixel ordinate location; if the gradation difference exceeds a set threshold value, a represents the a gray scale pixel coordinates where the ordinate mutation, and the ordinate and where the determined pixel abscissa point mutation as a reference point; calculating the point mutation, wherein the mutation point when the count reaches a when a predetermined value, the ordinate abscissa point mutation point and the point corresponding to the determined coordinate point as the solder joint.
  7. 7.根据权利5所述的识别方法,其特征在于,所述步骤(c)中,在以所述纵坐标点为基准的情况下,沿横坐标轴在纵坐标遍历范围内进行横坐标遍历,比较某一横坐标周围预定纵坐标范围(M2)内的横坐标所在像素与所述某一横坐标所在像素的灰度值;若灰度值的差超过设定阈值,则代表所述某一横坐标所在像素有灰度突变,并且该横坐标与所述作为基准的纵坐标点所确定所在像素为突变点;对所述突变点进行计算,其中,当所述突变点的计数到达某一预定值时,则将该突变点所对应的横坐标点与所述作为基准的纵坐标点确定为所述焊点的坐标点。 The identification method according to claim 5, wherein said step (c), the ordinate at the point in the reference case, the abscissa ordinate traversed within the range of traverse along the abscissa axis , where the gray value of a pixel within a relatively abscissa ordinate range around a predetermined horizontal axis (M2) and the abscissa where the certain pixel; if the gradation difference exceeds a set threshold value, a represents the a gray scale pixel where the abscissa mutation, and the abscissa the determined pixels located point mutation as ordinate reference point; calculating the point mutation, wherein, when the count reaches a point mutation when a predetermined value, then the corresponding point mutation point and the abscissa is determined as the coordinate point of the solder as ordinate reference point.
  8. 8.一种太阳能电池阵的焊接系统,其特征在于,包括: 图像采集模块,其用于采集需要焊接的银导线和银箔的图像; 图像识别模块,其接收所述图像采集模块所采集的采集图像,并识别出银导线焊接在银箔表面上的焊点,并且所述图像识别模块如前述权利要求1~7中任一项所述的图像识别装置; 焊接机构,其在所述焊点处对所述银导线和所述银箔进行焊接;及运动控制模块,其根据所述图像识别模块所识别出的焊点,控制所述焊接机构运动到所述焊点所在位置,完成所述焊点的焊接。 A solar array welding system, characterized by comprising: an image acquisition module that requires image acquisition soldering silver wire and silver foil is used; an image recognition module that receives said acquired image acquisition module image capture, and identify the silver wire bonding pads on the surface of the silver foil, and the image recognition module as claimed in the preceding image recognition apparatus as claimed in any one of claims 1 to 7; welding means, which in the welding at a point of the silver wire and the silver foil is welded; and motion control module, based on the image recognition module of the identified weld, said welding means is moved to control the position of pads is located, the complete the said weld pad.
  9. 9.根据权利8所述的焊接系统,其特征在于,所述图像采集模块包括低角度光源,以在采集时利用所述低角度光源对所·述银导线和银箔进行照明。 9. The welding system of claim 8, wherein the image acquisition module comprises a low angle light source, using said low angle light source is illuminated in the collection of the said silver-silver and lead.
CN 201210498499 2012-11-29 2012-11-29 Image recognition method CN103846539A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201210498499 CN103846539A (en) 2012-11-29 2012-11-29 Image recognition method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201210498499 CN103846539A (en) 2012-11-29 2012-11-29 Image recognition method

Publications (1)

Publication Number Publication Date
CN103846539A true true CN103846539A (en) 2014-06-11

Family

ID=50854918

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201210498499 CN103846539A (en) 2012-11-29 2012-11-29 Image recognition method

Country Status (1)

Country Link
CN (1) CN103846539A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104128709A (en) * 2014-06-30 2014-11-05 联合汽车电子有限公司 Automatic laser spot welding system and method based on vision-aided positioning

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040198858A1 (en) * 2002-12-20 2004-10-07 Brian Labrec Increasing thermal conductivity of host polymer used with laser engraving methods and compositions
CN101479566A (en) * 2005-11-14 2009-07-08 普雷茨特影像有限及两合公司 Method and device for assessing joins of workpieces
CN103810459A (en) * 2012-11-07 2014-05-21 上海航天设备制造总厂 Image recognition device and solar array?welding system by using same
CN103810458A (en) * 2012-11-07 2014-05-21 上海航天设备制造总厂 Image recognition method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040198858A1 (en) * 2002-12-20 2004-10-07 Brian Labrec Increasing thermal conductivity of host polymer used with laser engraving methods and compositions
CN101479566A (en) * 2005-11-14 2009-07-08 普雷茨特影像有限及两合公司 Method and device for assessing joins of workpieces
CN103810459A (en) * 2012-11-07 2014-05-21 上海航天设备制造总厂 Image recognition device and solar array?welding system by using same
CN103810458A (en) * 2012-11-07 2014-05-21 上海航天设备制造总厂 Image recognition method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104128709A (en) * 2014-06-30 2014-11-05 联合汽车电子有限公司 Automatic laser spot welding system and method based on vision-aided positioning
CN104128709B (en) * 2014-06-30 2016-04-27 联合汽车电子有限公司 Laser-based automatic spot welding system and method for locating the visual aids

Similar Documents

Publication Publication Date Title
US20080075385A1 (en) Detection and Correction of Flash Artifacts from Airborne Particulates
US20040247171A1 (en) Image processing method for appearance inspection
JP2009175334A (en) Microscopic system, image forming method, and program
CN101477687A (en) Checkerboard angle point detection process under complex background
US20120229618A1 (en) Defect inspection device and defect inspection method
US20080025616A1 (en) Fast multiple template matching using a shared correlation map
CN102692188A (en) Dynamic crack length measurement method for machine vision fatigue crack propagation test
US20100014759A1 (en) Eye detecting device, eye detecting method, and program
CN101957325A (en) Substation equipment appearance abnormality recognition method based on substation inspection robot
CN102393397A (en) System and method for detecting surface defects of magnetic shoe
CN101127083A (en) Welding image identification method
CN103116896A (en) Visual saliency model based automatic detecting and tracking method
CN102122390A (en) Method for detecting human body based on range image
CN102974551A (en) Machine vision-based method for detecting and sorting polycrystalline silicon solar energy
CN102175700A (en) Method for detecting welding seam segmentation and defects of digital X-ray images
CN102901444A (en) Method for detecting component size based on matching pursuit (MP) wavelet filtering and detecting system thereof
CN103136519A (en) Sight tracking and positioning method based on iris recognition
JP2013185862A (en) Defect inspection device and defect inspection method
US20120013727A1 (en) Cell characterization using multiple focus planes
US8315457B2 (en) System and method for performing multi-image training for pattern recognition and registration
CN101877074A (en) Tubercle bacillus target recognizing and counting algorithm based on diverse characteristics
CN103454285A (en) Transmission chain quality detection system based on machine vision
JP2011132010A (en) Wire rope inspecting device
JP2006337167A (en) Inspection method of low-contrast flaw under cyclic noise, and inspection method of low-contrast flaw under repeated pattern
He et al. Weld seam profile detection and feature point extraction for multi-pass route planning based on visual attention model

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination