201011311 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種影像檢測系統以及方法，尤指能以適 當光度進行檢測之影像檢測系統以及方法。 【先前技術】 在地球能源逐漸枯竭的今日，太陽能電池成為一代的新 眷 寵。以矽晶的太陽能電池而言，太陽能電池結構上包含層狀 相疊的光感應層以及背膠基板，光感應層表面上可見的為具 顏色的矽晶以及印刷在矽晶表面的導線。 導線顏色固定、均勻、且所佔面積較小，碎晶所佔面積 較大，由於矽晶的晶格不易控制，特別是多晶矽更為複雜， 所以即使同一批次所製成原本欲為同色的太陽能電池，往往 不同太陽能電池彼此間仍有色差，若不將其分類而任意組 裝，則外觀上可能是不同深淺的同色，看似斑駁而不美觀， ® 會影響到商品的價值與客戶的信賴感。 因此，在同色的太陽能電池出廠前須有分類的程序，該 站除了檢查瑕疵之外，最重要的就是將同色中屬性更接近的 同系顏色分為同一類，以便於後續安裝成的太陽能板較為美 觀。以往，這些工作都是以人工目視檢查，近年來已逐漸導 入自動光學檢測（Automated Optical Inspection ; A0I); 如中華民國專利第1282861號，於整個測試系統中，裝設有 光學檢測裝置，負責對太陽能電池的外觀進行檢測。 然而，即使現狀已逐漸導入自動光學檢測，但正因為石夕 201011311 bbJ色的不穩A ’ S線光度的控制不若過去檢測 印刷電路紐業、錢轉财麵耕制來的單純。若 不此以理想的光度來分檢分類太陽能電池的顏色，所分出的 顏色分類根本不是消費者所需要的。 ❹ 幻如、4費者於視覺上能接受為同一色系的兩片太陽能 電池、為晶格的關係，若使用相同光度的光線來照射，影 像感測器所感測出的灰階度可能差異很大而形成誤判；反 之’也可能使用相同歧的光線來照射，影像感測器所感測 出的灰階度很接近，但事實上消費者㈣卻是認定兩片 能電池不屬相同色系。 -因此，本發明的主要目的在於提供一種能以適當光度進 行檢測之影像檢n統以及方法，以改善上述問題。 【發明内容】 本發明之目的在提供一種能以適當光度進行檢測之影像 檢測系統錢方法，_冊每太雜電池，提供適當 光度的光線來檢齡敝陽能電池，使分_果既能符合^ 費者視覺的職，也使檢測過程自動快速而符合廠商之利益。 本發明係關於-種能以適當歧進行檢測之影像檢測系 統，係用以檢驗顏色符合一預定灰階度範圍之待測物。該影 像檢測系統係包含一光源設備、一影像感測器、一記憶單元、 以及一處理單元。 " 該光源設備係以一預定光度照射該待測物以產生反射 光。該影像感測器係接收反射光以測得一算數灰階度值。 6 201011311 雜理單元係接絲自郷像_狀算數灰階度 及擷取私憶單元巾之關聯方程式，該處料元將該算 階度值導人該關聯方程式中，以求取一操作光度。201011311 IX. DESCRIPTION OF THE INVENTION: TECHNICAL FIELD The present invention relates to an image detecting system and method, and more particularly to an image detecting system and method capable of detecting with appropriate luminosity. [Prior Art] Today, as the earth's energy is gradually depleted, solar cells have become a new generation of pets. In the case of twinned solar cells, the solar cell structure comprises a layered photo-sensing layer and a backing substrate. The surface of the photo-sensitive layer is visible with colored twins and wires printed on the twinned surface. The color of the wire is fixed, uniform, and the area occupied is small, and the area occupied by the crushed crystal is large. Since the crystal lattice of the twin crystal is difficult to control, especially the polycrystalline germanium is more complicated, even if the same batch is originally made to be the same color. Solar cells, often different solar cells still have color difference between each other. If they are not assembled and arbitrarily assembled, the appearance may be different shades of the same color, seemingly mottled and not beautiful, ® will affect the value of the goods and the trust of customers. sense. Therefore, in the same color solar cell, there must be a classification procedure before leaving the factory. In addition to checking the flaw, the most important thing is to classify the same color of the same color in the same color to facilitate the subsequent installation of the solar panel. . In the past, these tasks were all performed by manual visual inspection. In recent years, Automated Optical Inspection (A0I) has been introduced. For example, in the Republic of China Patent No. 1282861, optical detection devices are installed in the entire test system. The appearance of the solar cell is tested. However, even if the status quo has been gradually introduced into the automatic optical inspection, it is because of the control of the unstable A'S-line luminosity of Shixia 201011311 bbJ color, it is not as simple as the printing circuit New Year, the money to the financial sector. If the color of the classified solar cells is not classified by the ideal luminosity, the color classification is not required by the consumer at all.幻 幻幻, 4 fee users can visually accept two solar cells of the same color system, which is a lattice relationship. If the same illuminance is used to illuminate, the gray level difference sensed by the image sensor may be different. It is very large and misjudges; on the contrary, it may also use the same ray of light to illuminate. The gray level sensed by the image sensor is very close, but in fact, the consumer (4) believes that the two batteries are not of the same color. . - Therefore, it is a primary object of the present invention to provide an image inspection system and method capable of detecting with appropriate illuminance to improve the above problems. SUMMARY OF THE INVENTION The object of the present invention is to provide a method for image detection system that can detect with appropriate luminosity, and to provide a light of appropriate luminosity to detect the age of the solar cell, so that the score can be In line with the visual responsibility of the fee, the inspection process is automatically and quickly in line with the interests of the manufacturer. SUMMARY OF THE INVENTION The present invention relates to an image sensing system capable of detecting in an appropriate manner for testing a test object whose color conforms to a predetermined gray scale range. The image sensing system includes a light source device, an image sensor, a memory unit, and a processing unit. " The light source device illuminates the object to be tested with a predetermined illuminance to generate reflected light. The image sensor receives the reflected light to measure an arithmetic gray scale value. 6 201011311 杂 单元 系 系 _ _ _ _ _ _ _ 算 算 算 算 算 算 算 算 算 算 算 算 算 算 算 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联 关联Luminosity.
其^中’趟理單元係接續驅使該光源設備以該操作光度 =射4制物，再由該影佩卿透過反射光得到—檢測 j度值，當錄贼階度储合於歧魅範圍時， 该處理單元判定該待測物係屬核可。 因此’藉由本發明能以適當光度進行檢測之影像檢測系 :以及巧，利用處理單元所進行之演算手段，能夠針對每 片太陽i電池’提供適當絲的猶來檢浙類太陽能電 、'使々類結魏能符合、;肖#者視覺的細，也使檢測過程 自動快速而符合廠商之利益。 關於本發明之伽與精神可以藉由以下的㈣詳述及所 附圖式得到進一步的瞭解。 【實施方式】 :參閱第—圖’第—圖係本發明中太陽能電池1G之外觀 及側如思、圖。本發明所述制物之較佳標的係為太陽能電 展〇由側剖圖可見，太陽能電池1〇上、下係由一光感應 以及一背膠基板14疊合而成；以正視外觀圖可見，以 技〜方向正視光感應層12 ’係進一步包含-導線區域1202 以及-發晶區域腦，不論是多晶梦或是單晶結構，石夕晶區 201011311 域1204的顏色往往公欲丁^ 的將太陽能電_進行顏色的^相當技術手段才能自動 動化係=影像檢測系㈣所在自 中；自動化測試設備2β之前段係為進二 於了將太陽能電池1G進片於輸送帶21上之外，The 'mechanical unit' succeeds in driving the light source device to use the operation luminosity=shooting 4 artifacts, and then the shadow peeks are transmitted through the reflected light to detect the j-degree value, and when the thief grading is stored in the ambiguous range At the time, the processing unit determines that the object to be tested is approved. Therefore, the image detection system capable of detecting with appropriate illuminance by the present invention: and by using the calculation means performed by the processing unit, can provide appropriate silk for each solar cell of the solar cell, and The 々 结 魏 能 能 能 ; ; ; ; ; ; ; ; ; ; ; 肖 肖 肖 肖 肖 肖 肖 肖 肖 肖; The gamma and spirit of the present invention can be further understood by the following (d) detailed description and the accompanying drawings. [Embodiment]: Referring to Fig. 1 is a view showing the appearance and side of the solar cell 1G of the present invention. The preferred embodiment of the article of the present invention is a solar power display, which can be seen from a side cross-sectional view. The upper and lower sides of the solar cell are laminated by a light sensing and a backing substrate 14; In the direction of the technology, the light-sensing layer 12' further includes a wire region 1202 and a crystal region. Whether it is a polycrystalline dream or a single crystal structure, the color of the Shixiajing District 201011311 domain 1204 is often pleasing to the eye. The solar energy _ color control ^ equivalent technical means to automatically mobilize the system = image detection system (four) located in; the automatic test equipment 2β before the second section of the solar cell 1G into the conveyor belt 21 outer,
li背t 像檢喝統23，係照設並檢社陽能電池 10貪部用膠基板14表面之瑕疵。 册自純f試設備20之中段係為檢測主設備24,於輸送 π 21上方。又有影像檢測系統23，可用於檢測太陽能電池 光感應層12表面之顏色，根據所制得之色度值來對太陽能 電池10進行分類。 自動化測試設備2〇之後段係為分類設備26，於輸送帶 21之兩旁有許多的容置箱27，係將分類後之太陽能電池10 以機械手臂28運送而分置於不同的容置箱27中，供後續方 便出貨。 請參閱第三圖，第三圖係本發明影像檢測系統23之側面 刹示圖。本發明係關於一種能以適當光度進行檢測之影像檢 測系統23，係用以檢驗顏色符合一預定灰階度範圍之待測 物，可利用以分類如太陽能電池10之待測物的影像檢測系統 23 ’使後續能對太陽能電池1〇進行分檢分類。由第三圖可見 影像檢測系統23係包含一光源設備30、以及一影像感測器 32。 光源設備30係以一預定光度之光線照射如太陽能電池 8 201011311 10之待測物的光感應層12以產生一反射光。光源設備30進 一步包含一上光源模組3002、一下光源模組3004、以及一斜 光源模組3006 ;上光源模組3〇〇2以及下光源模組3004由環 狀分佈之多個LED燈30A構成，斜光源模組3006設於下光源 模組3004之下方，為四個四向之線光源，上光源模組3〇〇2 以及下光源模組3004對太陽能電池1〇產生正面光，斜光源 模組3006對太陽能電池10產生斜向光，藉由該等光源模組 來組合成所需之光線。 馨 影像感測器32可採用CCD影像感應器（CCD ImageLi back t like the inspection and drinking system 23, is set to check and check the social energy battery 10 greedy with the surface of the rubber substrate 14 瑕疵. The middle section of the pure test equipment 20 is the detection master device 24, which is above the transport π 21 . Further, an image detecting system 23 is provided for detecting the color of the surface of the solar cell photo-sensing layer 12, and classifying the solar cell 10 based on the obtained chromaticity value. The automatic test equipment 2 is a sorting device 26, and there are a plurality of accommodating boxes 27 on both sides of the conveyor belt 21, and the classified solar cells 10 are transported by the robot arm 28 and placed in different accommodating boxes 27 In order to facilitate subsequent shipments. Please refer to the third figure, which is a side view of the image sensing system 23 of the present invention. The present invention relates to an image detecting system 23 capable of detecting at an appropriate luminosity for detecting a test object whose color conforms to a predetermined gray scale range, and an image detecting system capable of classifying a test object such as a solar cell 10 23 'After the subsequent classification of solar cells can be classified. As seen from the third figure, the image detecting system 23 includes a light source device 30 and an image sensor 32. The light source device 30 illuminates the light-sensing layer 12 of the object to be tested, such as the solar cell 8 201011311, with a predetermined illuminance to generate a reflected light. The light source device 30 further includes an upper light source module 3002, a lower light source module 3004, and a oblique light source module 3006. The upper light source module 3〇〇2 and the lower light source module 3004 are distributed by a plurality of LED lamps 30A. The oblique light source module 3006 is disposed under the lower light source module 3004 and is a four-way line light source. The upper light source module 3〇〇2 and the lower light source module 3004 generate front light for the solar cell 1 The light source module 3006 generates oblique light to the solar cell 10, and the light modules are combined to form desired light.馨 Image sensor 32 can use CCD image sensor (CCD Image
Sensor)、CMOS 影像感應器（CMOS Image Sensor ; CIS)，係 設於光源設備30之上方。上光源模組3002以及下光源模組 3004環狀中間形成中空，影像感測器32透過此環狀中空接 收反射自下方太陽能電池1〇之反射光，以測得太陽能電池 10之影像畫面上複數個畫素之算數灰階度值。 配合第三圖進一步請參閱第四A圖，第四A圖係本發明 • 影像檢測系統23第一階段之功能方塊示意圖。影像檢測系統 23除了前述之光源設備3〇以及一影像感測器32之外，更包 含一記憶單元36以及一處理單元34。 記憶單元36係預先儲存一關聯方程式3602，其中關聯 方程式3602係為光度與灰階度之二元方程式。其中，為考量 精確度以及對硬體產生的負擔，經測試後，關聯方程式邪卯 係以二元二次方程式為佳。實務上，此二元二次方程式可由 多次實際的樣本太陽能電池10檢測，來測試出最佳參數，以 產生我們要的二元二次方程式。 9 201011311 處理單元34係接收來自影像感測器32之算數灰階度值 以^擷取記憶單元36中之關聯方程式36〇2，處理單元34將 δ玄算數灰度值導人關聯方程式删2巾，以求取一操作光 度。 配合第二圖、第四Α圖進-步請參閱第四Β圖，第四Β 圖係本發明影像檢測系統23第二階段之功能方塊示意圖。後 續，處理單S 34係接續驅使光源設備3〇以該操作光度再照A sensor, a CMOS image sensor (CIS), is disposed above the light source device 30. The upper light source module 3002 and the lower light source module 3004 form a hollow in the middle of the ring shape, and the image sensor 32 receives the reflected light reflected from the lower solar cell 1 through the annular hollow to measure the plurality of images on the solar cell 10 The grayscale value of the arithmetic of the pixels. Referring to the third figure, please refer to FIG. 4A, which is a functional block diagram of the first stage of the image detecting system 23. The image detecting system 23 further includes a memory unit 36 and a processing unit 34 in addition to the aforementioned light source device 3A and an image sensor 32. The memory unit 36 stores a correlation equation 3602 in advance, wherein the correlation equation 3602 is a binary equation of luminosity and gray scale. Among them, in order to consider the accuracy and the burden on the hardware, after testing, the associated equations are better than the binary quadratic equation. In practice, this binary quadratic equation can be tested by multiple actual sample solar cells 10 to test the optimal parameters to produce the binary quadratic equation we want. 9 201011311 The processing unit 34 receives the arithmetic grayscale value from the image sensor 32 to extract the correlation equation 36〇2 in the memory unit 36, and the processing unit 34 deletes the grayscale value of the δ meta-calculus correlation equation. Towels for an operational luminosity. Referring to the second diagram and the fourth diagram, please refer to the fourth diagram. The fourth diagram is a functional block diagram of the second stage of the image detection system 23 of the present invention. Subsequently, the processing single S 34 system continues to drive the light source device 3 to re-photograph the light with the operation.
射該待測物’再由影像感測器32透過反射光得到—檢測灰階 度值。 最後，處理單元34藉由該檢測灰階度值是否符合該預定 灰階度範圍來判定該待測物是不是屬於此分類範圍中；當該 檢測灰階度歸合_縣灰階晴，處理單元％ = 該，測物關核可；當該檢測灰階度值不符合於該預定灰階 度範圍時，處理單元34判定該制物係屬不核可。 請參閱第五圖’第五圖係本發明影像檢測方法之流程 =。本發明也係-魏料t歧進行檢敬影像檢測方 法’係用以檢驗顏色符合一預定灰階度範圍如太陽能電池⑺ 之待測物。該影像檢測方法係包含下列步驟： 利用反射光以 步驟S02 :以-預定光度照射該待測物， 測得一算數灰階度值。 :驟:將該算數灰階度值導入預儲之關聯方程式 度誠=之二元方程式。進一步，關聯方 7G二次方程式為佳。 201011311 步驟SG6 :以該操作光度再照射該待測物，利用反射光 以得到一檢測灰階度值。 步驟S07 :判斷該檢測灰階度值是否符合於該預定灰階 度範圍。 步驟S08 :當該檢測灰階度值符合於該預定灰階度範圍 時’則該待測物係屬核可。 步驟S10 .當該檢測灰階度值不符合於該預定灰階度範 ❿ 圍時’則該待測物係屬不核可。 因此，藉由本發明能以適當光度進行檢測之影像檢測系 統23以及方法，利用處理單元34所進行之演算手段，能夠 針$每一片太陽能電池10，提供適當光度的光線來檢測分類 太陽能電池10，使分類結果既能符合消費者視覺的預期，也 使檢測過程自動快速而符合廠商之利益。 藉由以上較佳具體實施例之詳述，係希望能更加清楚描 述本發明讀徵與精神麟以±述關露的較佳具體實 施例來對本發明之料加嫌制。相反地，其目的是希望能 涵蓋各觀變及具鱗性的安排於本發明所欲巾請之專利範 圍的範内。 【圖式簡單說明】 第-圖係本發明中太陽能電池之外觀及侧剖示意圖； _第二圖係本發明影像檢測系統所在自動化測試設備之 不意圖； 201011311 第三圖係本發明影像檢測系統之側面剖示圖； _係本發明影像檢測系統第-階段之功能方塊 圖； ϋ係本發明影像檢測系統第二階段之功能方塊 圍，Μ及 第五圖係本發明影像檢測方法之流程圖。The object to be tested is taken and then transmitted by the image sensor 32 to reflect the gray scale value. Finally, the processing unit 34 determines whether the object to be tested belongs to the classification range by detecting whether the gray scale value meets the predetermined gray scale range; when the detected gray scale is merged into the county gray scale, the processing is processed. The unit % = the measurement object can be checked; when the detected gray scale value does not conform to the predetermined gray scale range, the processing unit 34 determines that the system is not approved. Please refer to the fifth figure, the fifth figure, which is the flow of the image detecting method of the present invention. The present invention is also directed to a method for detecting images in which a color conforms to a predetermined gray scale range such as a solar cell (7). The image detecting method comprises the following steps: using the reflected light in step S02: illuminating the object to be tested with a predetermined illuminance, and measuring an arithmetic gray scale value. : Step: Import the arithmetic gray scale value into the pre-stored correlation equation. Further, the related party 7G quadratic equation is preferred. 201011311 Step SG6: The object to be tested is irradiated with the operation luminosity, and the reflected light is used to obtain a detected gray scale value. Step S07: It is judged whether the detected gray scale value conforms to the predetermined gray scale range. Step S08: When the detected gray scale value conforms to the predetermined gray scale range, the object to be tested is approved. Step S10. When the detected gray scale value does not conform to the predetermined gray scale range, the object to be tested is not approved. Therefore, the image detecting system 23 and the method capable of detecting with appropriate illuminance according to the present invention can use the calculation means performed by the processing unit 34 to supply the solar cell 10 with appropriate illuminance to detect the classified solar cell 10, The classification results can meet the expectations of the consumer's vision, and the detection process is automatically and quickly in line with the interests of the manufacturer. With the above detailed description of the preferred embodiments, it is intended to more clearly describe the preferred embodiments of the present invention and the spirit of the present invention. Rather, it is intended to cover a variety of variations and scales that are within the scope of the invention as claimed. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic view showing the appearance and side cross section of a solar cell in the present invention; _ second drawing is an intention of an automated test device in which the image detecting system of the present invention is located; 201011311 Fig. 3 is an image detecting system of the present invention FIG. 1 is a functional block diagram of the first stage of the image detecting system of the present invention; 功能 is a functional block of the second stage of the image detecting system of the present invention, and the fifth drawing is a flow chart of the image detecting method of the present invention. .
【主要元件符號說明】 太陽能電池10 背膠基板14 梦晶區域1204 影像檢測系統23 進片設備22 分類設備26 機械手臂28 影像感測器32 下光源模組3004 LED 燈 30Α 關聯方程式3602 光感應層12 導線區域1202 自動化測試設備20 輸送帶21 檢測主設備24 容置箱27 光源設備30 上光源模組3002 斜光源模組3006 記憶單元36 處理單元34[Main component symbol description] Solar cell 10 Backing substrate 14 Dream crystal area 1204 Image detection system 23 Film feeding device 22 Classification device 26 Robot arm 28 Image sensor 32 Lower light source module 3004 LED lamp 30 Α Correlation equation 3602 Light sensing layer 12 wire area 1202 automated test equipment 20 conveyor belt 21 detection main equipment 24 housing box 27 light source equipment 30 upper light source module 3002 oblique light source module 3006 memory unit 36 processing unit 34