TW200944747A

TW200944747A - Image measurement device and method for dimensional parameters of saw

Info

Publication number: TW200944747A
Application number: TW97113784A
Authority: TW
Inventors: Jui-Chang Lin; Yung-Cheng Wang; Shih-Fong Ciou; Hsiang-Yu Ho; Chieh-Cheng Chang
Original assignee: Univ Nat Formosa
Priority date: 2008-04-16
Filing date: 2008-04-16
Publication date: 2009-11-01
Also published as: TWI374252B

Abstract

The invention is concerned with the image measurement device for dimensional parameters of saw. It can be utilized for measurements of pitch, depth, radius of curvature and slope angle of saw. The basic equipments of this system are an image processing and analysis mechanism, human-machine control interface, a rigid stage, a clam installed on the stage, a image acquisition device fixed on the stage and connected with control device. A saw image of mounted on the clam can be gained by the image acquisition device and it can be transmitted to the control device. The original image can be transferred to the gray level image by the control device. Then the gray level image is transferred to binary image. After the process, coordinate of saw edge can be searched. With these gained data, the pitch, depth, radius of curvature and gap angle of saw can be calculated. Through the automatic measurement, the accuracy, efficiency and stable measurement results will be realized.

Description

200944747 九、發明說明：【發明所屬之技術領域】本發明係為一種檢測設備，尤指一種專門檢測鋸條尺寸的機具’以及其檢測方法。【先前技術】 φ 在處處講求品質、及生產力的今天，高生產力及最佳品質已是成形工廠中重要的課題。而要考慮生產力，品質的提昇，就必須正確地掌握住品質的檢驗的方式。目洳並無專門針對鋸條相關幾何尺寸之自動化檢測系統’因而無法提供鋸條加工機誤差修正之量化數據。而— 般以離線式（off_line)之人工（手動、目視）或機械檢測鑛條尺寸的方法’其量測誤差較大。另外對產品檢測速度亦有所限制’因此無㈣產品進行準確及有效率的檢測。【發明内容】 ▲有鑑於現有以人工或者機械檢測錯條尺寸的方法具有誤差大、效率差之缺失，士本發明之主要目的在於提供一種可以自動量測鋸條尺寸，裡並楗仏鋸條加工機對鋸條加工之口口貝判別及補償依據，以接古影像測量裝置及方法加工精度與生產效率的銀條基於上述目的，本發明的主要技術手尺寸參數之影像測量裝置，包括：故（、鑛條一控制裝置，為具右县彡 β 有景像處理運舁能力且具有人; 制介面的裝置；八令人機控200944747 IX. Description of the Invention: [Technical Field of the Invention] The present invention relates to a detecting device, and more particularly to a tool for specifically detecting the size of a saw blade and a detecting method thereof. [Prior Art] φ Every day, quality and productivity are everywhere. High productivity and best quality are important issues in the forming factory. In order to consider productivity and quality, we must correctly grasp the way of quality inspection. It is seen that there is no automated inspection system specifically for the geometry of the saw blade' and thus it is not possible to provide quantitative data for the error correction of the saw blade machine. In general, the method of manual (manual, visual) or mechanical inspection of the size of the ore strip in off-line has a large measurement error. In addition, there is a limit to the speed of product inspection. Therefore, no (4) products are accurately and efficiently tested. SUMMARY OF THE INVENTION ▲ In view of the existing method of manually or mechanically detecting the size of the wrong strip, which has the disadvantages of large error and poor efficiency, the main purpose of the present invention is to provide a saw blade processing machine capable of automatically measuring the size of the saw blade. According to the above objective, the image measuring device of the main technical hand size parameter of the present invention includes: The mine bar is a control device, which has the ability to handle and transport the image of the right county and has a human interface;

一固定平台，包括—A 基座、間隔設於基座頂面一背光 5 200944747 板與一移動座，背光板的板面是朝向移動座； -夾具，係設於基座上並且位於背光板與移動座之間’於夾具上可夾持待測的鋸條； —%<㈣取裝置’係設於㈣座上’其擷取影像的方向與鑛條表面垂直，該影像㈣裝置與㈣I置電性連接，且接受控制裝置之控制對待測的鑛條進行影像操取，亚將擷取結果傳送至控制裝置；鑤#中，該控制裝置依據所擷取的影像計算之尺寸參數。 1朱 —種鑛條尺寸參數之影像測量方法，其步驟包括·· 像；取得鑛條的影像：以-數位攝影機拍攝待測之鑛條影執行灰階化轉換：將掏取的影像經由灰階為灰階影像；得換以二元化轉換：將灰階影像進行二元化影像轉換， . ν成僅具有黑或白之二元化影像；搜尋鑛條影像邊緣之座標；里測及運算料尺寸減1所搜尋的座標進行運算、條之鋸齒的尺寸參數。為：實施上述技術手段以後，本發明可獲得的具體效益 1·本發明具備了自動定位量測或手動定錄條$去νη ν Μ ^ 功能U、齒根;罙、齒弧曲率半徑、間隙角等多項檢測 6 200944747 2-本發明之輸送裝置可延伸以進行線上;^、目,| 至鋸條加工機的生產線，仃線上榀測，以提升生產效率。 3.本發明利用影像分析原理，測結果。了獍传準確、穩疋的檢【實施方式】請參閱第一圖所千，盔姐鋸條(10)為且有甚” U、〇)的局部放大示意圖，的心側邊的扁平狀刀具，其中-側邊為平直锦m另—側邊設有多數個連續且間隔設置的尖錐狀接處尖錐端為齒尖（11Q)，二相鄰的㈣（11)的連的齒尖(11〇)二形的齒根(12) ’齒弧的二端分別與所鄰接即以一短切線(14)、-長切線(15)相切，本發明線測鑛齒(11)尖銳間隙角⑷〔即與平直的邊緣門 +線（16)與長切線（15)的夾角）、二鑛齒（11)之離、齒距（P)、齒根深（h)〔即齒尖⑴〇)至齒根（12)的距以及齒根（12)的齒弧曲率半徑（7 )。本發明鋸條尺寸參數之影像測量裝置有二種類型：一拾第，圖所示的線上檢測系、絶；二為第三圖所示的非線上 /糸統，其中，第二圖所示可線上檢測的影像測量裝置 4含-控制裝置(20)、一固定平台(30)、一夾具(40)以及 /像㈣裝置（5G)，H所述的控制裝置（20)為且有衫像處理運算能力且具有人機控制介面的裝置，其可以為人電腦或為—控制^，可執行—鑛條尺寸參數之影、涮置方法，本較佳實例之控制裝置（2〇)係為—個人電腦，括相互電性連接的一主機、一影示器以及一輸入單元， 200944747 其中，該主機設有與其他電控設備（如伺服馬達' ccD、 CMOS·等）連接之電氣訊號輸出入介面。 ❹ 所述固疋平台（30)包括一基座（31 )、一輸送裝置（32)、一移動座（33)以及一背光板（34)，該輸送裝置（32)可於基座 (31)頂面滑移並且延伸至一鋸條加工機（6〇)，輸送裝置（32) 包括一承載單元（320)以及一驅動承载單元（32〇)位移的伺服馬達（圖中未示），該承載單元（32〇)可以是底部設有一滑軌裝置的平台、亦可以為一輸送帶，該伺服馬達與控制裝置（20)電性連接，並接受控制裝置（2〇)的之控制而轉動至預定位置。所述的移動座（33)及背光板（34)係設於基座（31) 頂面並且分別位於輸送裝置（32)二側，該移動座（33)係以可位移狀叹於基座（3彳）頂面，且其移動方向與輸送裝置Ο?) 的移動方向垂直，該移動座（33)上並且設有一上下延伸的調整槽（331)，所述背光板（34)為可發出面光源的平板狀背光板，其固定於基座（31)頂面並且板面朝向輸送裝置02) 及移動座（33)，較佳設為白光背光板。所述夾具（40)係設於輸送裝置（32)頂部，其可以夾持錯條（1〇)，並且㈣條⑽的鑛齒朝上，失具（4g)夾持的鑛條（1 〇)表面係與移動座（33)的位移方向垂直。所述的影像擷取裝置（50)與該控制裝置（20)連接，其 :可位移狀設於移動座(33)的調整槽(33”上，影像擷取裝 _係接受控制裝置(2〇)之控制而進行影像掏取並將擷取結果持續傳送予兮·於告,丨驻丨川彳。太y a 取梦…… 佳實施例之影像擷、置（）匕括一數位攝影機（51)(較佳為CCD)，以及一 8 200944747 接設於數位攝影機^ 51)夕旦，& 陣朝向夾二))讀取端的鏡頭（52)，且鏡頭 $ #由承載單7(32G)將夹具（40)輸送至鑛條加 Ο 工機_之生產線，將加工完成的鑛條放置於央且_)上，以飼服馬達驅動使承載單元⑽）朝向基座（Μ)位移，至央之^連同錯條（1G)位移至影像㈣裝置（5G)與背光板（34) :止’而後，控制裝置（2G)即開始執行㈣條尺寸參 ::像测量方法，控制裝置(2〇)控制數位攝影機(5”操 H ’而該背光板㈣提供白光背景使操取結果更為清折，所擷取的錯條(10)影像透過數位攝影機(5”内部的一擁取卡傳輸到控制I w ^ 裝置（〇) ’經由控制裝置（20)將操取結半徑、二【像處理’再作蘇條的齒距、齒根深、齒弧曲率工佟正㉟角等夕項檢測功能，提供鋸條加工機(60)再加 ^正鋸條參數之依據。 ❹ 巧參閱第三圖所示，所述非線上檢測系流的鋸條尺寸錄條2 1像心裝置’係獨立於錫條加工機以外’而不與 a(3(TA，機連接’其仍然包括—控制裝置（2〇)、—固定平 :平ΙΟ二Ϊ具(4〇)、以及一影像擷取裝置(5〇)，而該固座(3二及;J座(31)、一具有調整槽一了貝J 責光板(34)，所述夾具(4〇)係固定基座(31) 裝置⑽^由^光板Μ與移動座（33)之間，而影像操取 ⑸)心· Γ:機(51)與鏡頭(52)組成，數位攝影機第二圖:==整：!31)上，，相較於前述 jir、統而S，本實施例不需要輸送裴置（32) 9 200944747 與Μ條加工播遠全士 ^ 栻連軋，其餘檢測方式均相同。使用時，將完成加工的鋸條（ ;進行_分析不良…另—方面可作為加工可以檢置（2〇)的軟體者中& λ ^ 此外，.可在控制裝瓶田r 5又疋鋸條角度、長自動判斷待測物件 "乾圍，糟由電腦 ^, w物件之優名，而使產品升產品的品質、良率。動進仃/刀類，以提月 > 閱第四圖所示，該錯條的步驟包括： ’、、4數之影像測量方法取得鋸條的影像··控制裝置（2〇取得所拍攝待測錯條的影像; “象掏取裝置(5〇) 為灰：:階化轉換:將操取的影像經由灰階運算，轉換形成= 將灰階影像再度經過二元化處理，、或之—元化影像，其中，-_ il· ❹ 式可以為-單—閥值-Μ轉換的方值），依昭每個禮“先在原始I像中設定灰階值々（閥轉換為里色4 ^ 於次小於將像素值為255)，若灰ptb估^ 象素轉換為白色（灰階以取得今待、,& ，則像素轉換為黑色（灰階值為0)，女係以〇tSU演算法獲得；佳實施例之該閥值搜哥鋸條影像邊緣之座標：搜尋〔如第—圄张_ ’條衫像邊緣之座標 !”所不之二齒尖(11〇)位置，即所標示座標a、。；里測及運异鋸條尺寸參#支： !仪辱的座標a、座標f 10 200944747 進仃運异錯條之齒距（p )、齒根深（h )、齒弧間隙角（Θ )箅赵诚卞干L ( r ) 數據，並藉由線性化、平均化作誤以降低量測之誤差。、是仏正據提:=上檢測時，將所得到的各項檢測項目參數之數 IT:::加工機，以修正加工數據，再藉此數值建立測：=控管之具體定量分析依據；倘為非線上檢 _ ❹ 及修正加工之;:f判斷及控管之具體定量分析依據，以本發明『鑛條尺寸參數之影像檢測方法』是以一所：制裝f :中：軟體「鋸條尺寸參數之影像檢測系統」用ci 式是使用高階程式語言軟體撰寫，其方式利盘數位私2抑影像揭取裝置掏取影像，以幾何量測原理 =位^處理法作鑛條之檢測。先由ccDca_a操取 h 取到㈣像作灰階運算，將影像轉換為灰階值/象隹再利用otsu演算法（otsu，s met = 閥值運算’以將影像轉換為二元(黑…數位 =，再搜哥錯條影像邊緣之座標來運算 =齒弧曲率半徑、間隙角等數據，並藉由線性化、；句匕作块差修JL，其軟體程式介紹如下·· 1 ·單一閥值運算马像Γ皆影像要轉為二元（黑白）數位影像，可以先在原始 =設定灰階值々，依照每個像素的灰階值大小(大於或 J、於）’將像素轉換為黑色或白色： —A fixed platform includes an -A base, a backlight disposed on the top surface of the base, a backlight 5 200944747 board and a moving base, the surface of the backlight board is facing the moving seat; - a fixture, which is disposed on the base and located in the backlight The saw blade to be tested can be clamped on the jig with the moving seat; -%<(4) the device is mounted on the (four) seat's direction of the image being perpendicular to the surface of the bar, the image (4) device and (4) I The electrical connection is connected, and the control device controls the image of the ore to be tested, and the result of the extraction is transmitted to the control device; in 鑤#, the control device calculates the size parameter according to the captured image. 1 Zhu----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- The order is a grayscale image; it can be converted into a binary transformation: the grayscale image is converted into a binary image, ν is a binary image with only black or white; the coordinates of the edge of the image of the mineral bar are searched; The size of the calculation material is reduced by 1 and the coordinates of the coordinates of the bar are calculated. After the implementation of the above technical means, the specific benefits obtainable by the present invention are as follows: 1. The invention has an automatic positioning measurement or manual recording strip $ deνη ν Μ ^ function U, root; 罙, tooth radius of curvature, clearance Angle and other multi-detection 6 200944747 2- The conveying device of the present invention can be extended to carry out on-line; ^, mesh, | to the production line of the saw blade processing machine, and on the 仃 line to improve production efficiency. 3. The present invention utilizes the principle of image analysis to measure results.獍准确准确准确准确准确准确实施实施实施准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确准确请请请请请请请请请请请请请请请Wherein - the side is flat and the other side is provided with a plurality of continuous and spaced tapered end joints which are tip ends (11Q) and two adjacent (four) (11) connected tips. (11〇) Two-shaped root (12) The two ends of the tooth arc are respectively tangent to a short tangent (14) and a long tangent (15), and the wire measuring tooth (11) of the present invention is sharp. Gap angle (4) [ie, the angle between the edge door + line (16) and the long tangent line (15), the separation of the second tooth (11), the pitch (P), the root depth (h) [ie the tip (1) 〇) the distance from the root (12) and the radius of curvature of the tooth root (12) (7). There are two types of image measuring devices for the size parameter of the saw blade: one pick, the online test shown in the figure The second embodiment is a non-line/system according to the third figure, wherein the image measuring device 4 capable of on-line detection shown in the second figure includes a control device (20), a fixed platform (30), and a fixture. (40) and / like (4) Device (5G), the control device (20) described in H is a device having a processing capability of a shirt image and having a human-machine control interface, which may be a human computer or a control, executable - size parameter of the ore bar The control device (2〇) of the preferred embodiment is a personal computer, including a host, a video display, and an input unit electrically connected to each other, 200944747, wherein the host is provided with Other electrical control devices (such as servo motor 'ccD, CMOS, etc.) are connected to the electrical signal input and output interface. ❹ The fixed platform (30) includes a base (31), a conveying device (32), and a moving seat. (33) and a backlight (34), the conveying device (32) is slidable on a top surface of the base (31) and extends to a saw blade processing machine (6〇), and the conveying device (32) includes a carrying unit ( 320) and a servo motor (not shown) for driving the carrying unit (32〇), the carrying unit (32〇) may be a platform with a sliding rail device at the bottom, or a conveyor belt, the servo motor Electrically connected to the control device (20) and accept control (2〇) is rotated to a predetermined position. The moving seat (33) and the backlight (34) are disposed on the top surface of the base (31) and are respectively located on two sides of the conveying device (32), the movement The seat (33) is slidably slid on the top surface of the base (3彳), and its moving direction is perpendicular to the moving direction of the conveying device ,?), and the moving seat (33) is provided with an adjusting groove extending upward and downward. (331), the backlight (34) is a flat backlight capable of emitting a surface light source, which is fixed on the top surface of the base (31) and faces the conveying device 02) and the moving seat (33), preferably It is a white light backlight board. The clamp (40) is attached to the top of the conveying device (32), which can clamp the wrong strip (1〇), and (4) the mineral tooth of the strip (10) faces upward, and the metal strip (4g) is clamped (1 〇) The surface system is perpendicular to the displacement direction of the moving seat (33). The image capturing device (50) is connected to the control device (20), and is displaceably disposed on the adjusting slot (33) of the moving base (33), and the image capturing device is received by the control device (2) 〇) control and image capture and continuous transmission of the results to 兮·于告,丨在丨川彳.太雅的梦...... The image of the best example, set () includes a digital camera ( 51) (preferably CCD), and a 8 200944747 connected to the digital camera ^ 51) Xidan, & array facing clip 2)) read end lens (52), and lens $ # by carrier single 7 (32G The jig (40) is transported to the production line of the miner twisting machine _, and the processed ore is placed on the center _), and the load bearing unit (10) is displaced toward the pedestal (Μ) by the feeding motor drive. To the central ^ together with the wrong strip (1G) displacement to the image (four) device (5G) and the backlight (34): stop 'and then, the control device (2G) begins to execute (four) size parameters:: image measurement method, control device ( 2〇) Control the digital camera (5" H" and the backlight (4) provides a white background to make the results of the operation more clear, the wrong (10) image is read through The digital camera (5" inside a card is transferred to the control I w ^ device (〇) 'The control device (20) will operate the junction radius, two [image processing] and then the pitch of the strip, the root depth, The tooth arc curvature process is the 35th angle and other eve detection function, and provides the basis of the saw blade processing machine (60) plus the positive saw blade parameter. ❹ Referring to the third figure, the saw blade size record of the non-line detection system flow Article 2 1 The image heart device 'is independent of the tin processing machine' and not with a (3 (TA, machine connection 'it still includes - control device (2〇), - fixed flat: flat two cookware (4 〇), and an image capture device (5〇), and the fixed seat (3 2 and; J seat (31), one with an adjustment slot, a shell J light plate (34), the fixture (4〇) The fixed base (31) device (10) ^ is between the light plate Μ and the moving seat (33), and the image manipulation (5)) heart Γ: the machine (51) and the lens (52), the digital camera second picture: == 整:!31), compared with the above jir, system S, this embodiment does not need to transport the device (32) 9 200944747 and the processing of the smashing machine, the full squadron ^ 栻 rolling, the rest of the detection methods are the same When used, the saw blade will be processed (; _ analysis is bad... another aspect can be used as a software that can be processed (2〇)& λ ^ In addition, can be controlled in the bottling field r 5 and 疋 saw blade Angle and length automatically judge the object to be tested "dry circumference, the bad name of the computer ^, w object, and make the product quality and yield of the product. Move into the knife / knife, to lift the moon > Read the fourth As shown in the figure, the steps of the stagger include: ', 4 image acquisition methods to obtain the image of the saw blade · · Control device (2 〇 to obtain the image of the wrong strip to be taken;; icon capture device (5 〇) Gray:: gradation conversion: the image to be manipulated is converted into a gray-scale operation, and the gray-scale image is again subjected to binarization processing, or a meta-image, wherein -_ il· ❹ can be - Single - Threshold - 方 conversion square value), according to each ceremony "first set the gray level value in the original I image 々 (the valve is converted to the inner color 4 ^ the second time is less than the pixel value is 255), if gray Ptb estimates ^ pixels are converted to white (gray order to get the current, &, then the pixel is converted to black (grayscale value is 0) The female department is obtained by the 〇tSU algorithm; the coordinates of the threshold of the image of the best example of the search saw blade: search (such as the first - 圄 _ 'striped figure like the edge of the edge!" ) position, that is, the coordinates a,.里里里里里里锯锯 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 Zhao Cheng dries the L ( r ) data and uses linearization and averaging to reduce the error of the measurement. It is said that: ● On the test, the number of parameters of each test item obtained will be IT::: processing machine to correct the processing data, and then the value is used to establish the test: = the specific quantitative analysis basis of the control If it is a non-online inspection _ ❹ and correction processing; : f judgment and control specific quantitative analysis basis, the invention "mineral size parameter image detection method" is one: manufacturing f: medium: software "Image detection system for saw blade size parameters" is written using high-level programming language software. The method is to use the digital display method to capture images. The geometric measurement principle = bit ^ processing method for mining. Detection. First, ccDca_a fetches h (4) as gray-scale operation, converts the image to grayscale value/symbol and then uses otsu algorithm (otsu, s met = threshold operation) to convert the image into binary (black... Digit =, then search for the coordinates of the edge of the image of the wrong image to calculate = tooth radius curvature, clearance angle and other data, and by linearization, sentence block for the block repair JL, the software program is introduced as follows · · · Single Threshold operation horse image is to be converted into binary (black and white) digital image, you can first convert the pixel according to the grayscale value of each pixel (greater than or J, in) in the original = set grayscale value Black or white: —

Xh值則像素轉換為白色（灰階值為255) 200944747 灰階值以，則像素轉換為黑色（灰階值為ο) 〇tsu /貝算法（Otsu’s method) 將’?、_/像值方圖視為一種機率分布The Xh value converts the pixel to white (the grayscale value is 255). 200944747 The grayscale value is, then the pixel is converted to black (the grayscale value is ο) 〇tsu / Bayesian algorithm (Otsu's method) will be '?, _ / image value side Graph as a probability distribution

Pi = «,·/// € :像素灰階值為/的機率 '•灰階值為Μ象素數目 ^ :像素總數若閥值設為λ，則定義 /=〇Pi = «,·/// € : probability of pixel grayscale value of / '• grayscale value Μ pixel number ^ : total number of pixels If the threshold is set to λ, then define /=〇

响=IU 其中1為1灰階數（1-1則為最大灰階數）。按照定義計算出蚴)與咐）差異最大時的平均為 uT^^ipi 可以先定義影像Ring = IU where 1 is the 1 gray scale (1-1 is the maximum gray scale). According to the definition, the average difference between 蚴) and 咐) is uT^^ipi. You can define the image first.

找出下列函數值最大的灸值 Hk)u(k) 3 ·鑛條之齒距量測利用單一閥值運算與〇tsu演算 4 ji . a*： ^ 、士鋸條影像做影像將衫像轉換為二元（黑白）數位影像輪廓邊緣之应戶a盥1像’再搜哥鑛條影 1遭、，象之座私a與座標f(如第一圖條之齒距。齒大）來運算錄Find the moxibustion value of the following function value Hk)u(k) 3 · The pitch measurement of the ore bar uses a single threshold value calculation and 〇tsu calculus 4 ji . a*: ^, the saw blade image is used to image the shirt image conversion For the binary (black and white) digital image contour edge of the household a盥1 like 'research of the mine shadow 1 is, the image of the private a and the coordinates f (such as the first bar pitch. Tooth large) come Calculation record

鋸條之齒根深量測 12 200944747 们用早—閥值運算與〇tSU演曾處理，將影像轉換為二元(黑白)數：影像錯條影像做影像像邊緣之齒弧頂點座標G與座標再搜尋鑛條影之齒根深。圖）來運算鎇條 ❹ φ 5·鋸條之齒弧曲率半徑量測利用單一閥值運算與〇tsu演算法處理’將影像轉換為二元(黑白)數位影像：再像像邊緣之座標，如第—圖之齒根（12)的齒去，與長切線05)的相切“、以及齒弧上的_點二齒弧 …的距離，等於。點至b點的距離，而,點二^^ 少】’ C點座標為七、少2，d點座標為七、少3，分別將卜d 三點的座標…、m、乃代入以下所示之圓方程式 (1)、（2)、（3)，三個聯立方程式可求得三個未知係數c、d、 E，將求得的係數C、D、E帶入式（4)可求得齒弧曲率半徑。圓的方程式(乂-乂0)2+(产)/0)2=#可化成二元二次方程式x2+y2+Cx+Dy+E=〇的形式，如式(1)、（2)、（3)。The root depth measurement of the saw blade 12 200944747 We used the early-threshold calculation and the 〇tSU performance to convert the image into binary (black and white) numbers: the image is staggered and the image is like the edge of the tooth. Search for the root depth of the mine shadow. Fig.) To calculate the radius of the ❹ 5 φ 5 · saw blade curvature radius measurement using a single threshold operation and 〇 tsu algorithm processing 'convert the image into a binary (black and white) digital image: then like the edge of the image, such as The tooth of the tooth root (12) of the first figure, the tangent to the long tangent line 05), and the distance of the _ point two tooth arc of the tooth arc are equal to the distance from the point to the point b, and the point two ^^ Less] 'C coordinates are seven, less 2, d coordinates are seven, less three, respectively, the coordinates of the three points of the d..., m, is substituted into the circular equations (1), (2) shown below (3), three simultaneous equations can obtain three unknown coefficients c, d, E, and the obtained coefficients C, D, E can be obtained into equation (4) to obtain the radius of curvature of the tooth arc. (乂-乂0)2+(production)/0)2=# can be converted into a binary quadratic equation x2+y2+Cx+Dy+E=〇, as in equations (1), (2), (3) ).

Cx] +Dyx+E = -(λ：,2 + y]) ............................⑴Cx] +Dyx+E = -(λ:,2 + y]) ............................(1)

Cx2 + Dy2 +E = -(xl + ) ............................(2)Cx2 + Dy2 +E = -(xl + ) ............................(2)

Cx3 +Dy3+E = ~(x23 + y]) ............................⑶ Χ2+>^+〇Χ+ϋ)/+Ε=0 =>(x+^)2+(y+^)2_^_+^__^ 齒弧曲率半徑+ --£ V 4 4 13 (4) 200944747 6.錯條之間隙角量測線性化，最小平方法. 取得同-長切線（15)上多對义的線性迴歸線為、上標（ά)，“·，(ά)，設定少 II Σ (乃-(办+似,.))2 >=1 使得〔最二法求出趣歸係數6、 „ 最小。我們可鯉山、 J解出迴歸係數分別 a = —_______ ^ Σ^-χ)2 /-1Cx3 +Dy3+E = ~(x23 + y]) ......................(3) Χ2+>^+〇Χ +ϋ)/+Ε=0=>(x+^)2+(y+^)2_^_+^__^ Tooth Curvature Radius + --£ V 4 4 13 (4) 200944747 6. Staggered gap Angle measurement linearization, minimum flat method. Obtain the linear regression line of multiple pairs on the same-long tangent line (15), superscript (ά), "·, (ά), set less II Σ (乃-(办+ Like, .)) 2 > = 1 makes [the second method to find the interesting coefficient 6, „ minimum. We can solve the regression coefficients of Lushan and J respectively. a = —_______ ^ Σ^-χ)2 /-1

A T — Λ — b = y~ax 其中 1 η π丨係數ά即為第—圖中的紂 ©其間隙角如-,u)。、、又[即長切線（15)J的斜率【圖式簡單說明j 第圖係本發明所欲檢測第二圓係本發量裝置。之鋸條的局部放大示意圖明線上檢測型的鋸條尺寸參數之影像測第二圖係本發明非測量裝置。線上檢測型的錯條尺寸參數之影像第四圖係本發明鋸條尺寸參數〜像測置方法的流程 ]4 200944747 【主要元件符號說明】 (10)鑛條 (11)鑛齒 (110)齒尖 (1 2)齒根 (14)短切線 (1 5)長切線 (16)水平線 (20)控制裝置 (30)(30A )固定平台 (31)基座 (32)輸送裝置 (320)承載單元 (3 3)移動座 (331)調整槽 (3 4)背光板 (40)夾具 (50)影像擷取裝置 (51)數位攝影機 (52)鏡頭 (6 0)鑛條加工機 (P )齒距 (h )齒根深 (r )曲率半徑 (0 )間隙角。 15A T — Λ — b = y~ax where 1 η π丨 The coefficient ά is the 纣 in the graph—the gap angle is -, u). And [that is, the slope of the long tangent line (15) J. [Fig. 1 is a schematic diagram of the second round system of the present invention. A partially enlarged schematic view of a saw blade. An image measurement of a saw blade size parameter on a line detection type. The second diagram is a non-measurement device of the present invention. The fourth image of the linear detection parameter of the line detection type is the flow of the saw blade size parameter to the image measurement method of the present invention. 4 200944747 [Explanation of main component symbols] (10) Mineral bar (11) ore tooth (110) tooth tip (1 2) Tooth root (14) Short tangent (1 5) Long tangent (16) Horizontal line (20) Control device (30) (30A) Fixed platform (31) Base (32) Conveying device (320) Carrying unit ( 3 3) Mobile seat (331) adjustment slot (3 4) backlight board (40) fixture (50) image capture device (51) digital camera (52) lens (60) ore processing machine (P) pitch ( h) Root depth (r) radius of curvature (0) clearance angle. 15

Claims

200944747 十、申請專利範圍： 1 . 一種鋸條尺寸參數之影像測量裝置，包括：一控制裝置’為具有影像處理運算能力且具有人機押制介面的裝置；工一固疋平台，包括一基座、間隔設於基座頂面一背光板與一移動座’背光板的板面是朝向移動座；一夹具，係設於基座上並且位於背光板與移動座之間’於夹具上可夾持待測的鋸條；參一影像擷取裝置，係設於移動座上，其擷取影像的方向與鋸條表面垂直’該影像擷取裝置與控制裝置電性連接，且接受控制裝置之控制對待測的鋸條進行影像擷取，並將擷取結果傳送至控制裝置；其中’該控制裝置依據所擷取的影像計算該待測鑛條之鑛齒尺寸參數。 2 如申清專利範圍第1項所述之鑛條尺寸參數之影像測量裝置，其中，所述的移動座上設有一可供影像擷取鬱裝置升降位移的調整槽。 3 .如申請專利範圍第2項所述之鋸條尺寸參數之影像測量裝置，其中，所述的移動座係以可朝向背光板方向位移狀設於基座上。 4 .如申請專利範圍第1或2或3項所述之鋸條尺寸參數之影像測量裝置，其中，基座頂面設有一位於背光板與移動座之間的輸送裝f ’其包括一承載單元以及一驅動承載單元之伺服馬達，所述的夾具係5又於承載單元上，該 16 200944747 飼服馬達伯仿3 4 ” 4制裝置電性連接，而驅動承載單开## . 且接又控制裝置之控制 w . 夕，承载單元的位移方向盥夯像m # 置之擷取影像方向垂直。〃知像擷取裝 5 .如申請專利範圍第4項所像測量裝置，、鋸條尺寸參數之影機、以及-g裝置包括-數位攝影接於數位攝影機的一鏡頭。 6.-種鋸條尺寸參數之影像測量方法 ❹200944747 X. Patent application scope: 1. An image measuring device for saw blade size parameters, comprising: a control device as a device having image processing computing capability and having a human-machined interface; a work-solid platform comprising a base a plate disposed on the top surface of the base and a backlight of the base plate is a moving seat; a clamp is disposed on the base and located between the backlight and the movable seat. Holding the saw blade to be tested; the image capturing device is disposed on the moving seat, and the direction of capturing the image is perpendicular to the surface of the saw blade. The image capturing device is electrically connected to the control device, and is controlled by the control device. The saw blade performs image capturing, and transmits the captured result to the control device; wherein the control device calculates the ore tooth size parameter of the to-be-measured ore according to the captured image. 2 The image measuring device of the size parameter of the ore strip according to claim 1, wherein the moving seat is provided with an adjusting groove for the lifting and lowering displacement of the image capturing device. 3. The image measuring apparatus of the blade size parameter according to claim 2, wherein the moving seat is disposed on the base so as to be displaceable toward the backlight. 4. The image measuring device of the saw blade size parameter according to claim 1 or 2 or 3, wherein the top surface of the base is provided with a transport device f' between the backlight plate and the movable seat, which comprises a carrying unit And a servo motor for driving the load bearing unit, wherein the clamp system 5 is further mounted on the load bearing unit, and the 16 200944747 feed clothes motor-like motor is electrically connected, and the drive carries a single open ##. Control of the control device w. In the evening, the displacement direction of the carrying unit is like the m # setting and the image direction is perpendicular. 〃撷撷 . . . . . . . . . . . . . . . . . . . . . 如如如如如The camera and the -g device include a lens for digital photography connected to a digital camera. 6.-Image measurement method for the size parameter of the saw blade❹

取得鋸條的县/总. …/驟包括. 像； …影像·以一數位攝影機拍攝待測之鑛條影執行灰階化轉換為灰階影像；將擷取的影像經由灰階運算，轉換以r 2二元化轉換：將灰階影像進行二^影像轉換，形成僅具有黑或白之二元化影像，· 搜尋鋸條影像邊緣之座標；量測及運算鑛條尺寸參數：以所搜尋的座標進行運算鑛條之鋸齒尺寸參數。、7 ·如申請專利範圍第6項所述之鋸條尺寸參數之影、、J里方法其中，所述鋸條的一側設有一平直的邊緣線，山側n又有夕數個連續且間隔設置的尖錐狀鋸齒，其尖錐 =為齒尖’二相鄰的鑛齒的連接處形成—圓狐形的齒根，齒根的齒弧二端分別與所鄰接的二齒尖以一短切線、—長切線相切，齒尖與所述邊緣線平行的延伸線設為水平線，該水平線與長切線之間構成齒尖的間隙角，本發明為檢測鋸齒的齒距、齒根深、齒弧曲率半徑、間隙角等數據。 17 200944747 8 ·如申請專利範圍第7項所述之鋸條尺寸參數之，像測量方法’其中，量測齒距的方法係搜尋鋸條影像輪廊邊緣之二齒尖座標，以運算齒距。 9 ·如申請專利範圍第8項所述之鋸條尺寸參數之貪像測量方法，其中，量測齒根深的方法係搜尋鋸條影像邊緣之齒弧頂點座標與鄰接之齒尖座標，以運算齒根深。 10.如申請專利範圍第9項所述之鋸條尺寸參數之影像’則置方法，其中，量測齒弧曲率半徑的方法係搜尋鋸條影像邊緣之座標，將齒根的齒弧三個點座標^、、 A、h代入以下圓方程式（1)、（2)、（3)，三個聯立方程式可求得三個未知係數C、D、E，將求得的係數c、d、e帶入式（4)可求得齒弧曲率半徑：圓的方程雄-砧，，？可化紅元二対料心+Dy+E=〇的形式，如式(1)、（2)' (3): ⑴ .(2) .(3) Cx, +Dy}+E = -(xf + yf) Cx2 + Dy2 + E = -{xl + j Cx3 + Dy3 +E = -(¾- + ^>32) x2V+Cx+Dy+E=0 =>(x+f )2+(y+|)2=^+^E V 4 4 .......................{V 1 1 .如申請專利範圍第9 貝所迷之鋸條尺寸參數之影像測量方法，其中，鋸條之間隙角量測： 18 200944747 取得同一長切線上多點的座標（X,，)，,)，...,(u„)，設定y對X 的線性迴歸線為 y-b-v αχ 以最小平方法求出迴歸係數&、α，使得 Σ〇;,· -φ + αχ:))2 ί=ι 最小。解出迴歸係數分別為 a=J^L—n- ^(x,· -x)2 /=1 b = y - αχ 其中 1 « ^=-Σ^· η /=ι 歹=-1^兄· η /=ι 其間隙角為θ = tan 1 ( 5)。十一、圖式：如次頁 19The county/total of the saw blade is included in the ..../image includes: image; ... image taken by a digital camera to perform grayscale conversion to grayscale image; r 2 binary conversion: the gray image is converted into two images, forming a binary image with only black or white, · searching for the coordinates of the edge of the saw blade image; measuring and calculating the size parameter of the ore bar: searching for The coordinate is used to calculate the sawtooth size parameter of the ore. 7. The shadow size parameter of the saw blade according to item 6 of the patent application scope, wherein the side of the saw blade is provided with a straight edge line, and the mountain side n has a plurality of consecutive and spaced intervals. The tapered taper tooth has a pointed cone = a toothed tip formed at the junction of two adjacent orthodontic teeth - a rounded fox-shaped tooth root, and the two ends of the toothed root of the tooth root are respectively shorted with the adjacent two-toothed tip The tangent line, the long tangent line is tangent, and the extension line parallel to the edge line is set as a horizontal line, and the gap line between the horizontal line and the long tangent line forms a gap angle of the tooth tip. The present invention detects the pitch of the serration, the root depth, and the tooth. Arc curvature radius, clearance angle and other data. 17 200944747 8 • As for the saw blade size parameter described in item 7 of the patent application, the method of measuring the pitch, wherein the method of measuring the pitch is to search for the two-toothed tip coordinates of the edge of the saw blade image gallery to calculate the pitch. 9. The image measuring method of the saw blade size parameter according to claim 8, wherein the method of measuring the tooth root depth searches for the tooth arc vertex coordinates of the edge of the saw blade image and the adjacent tooth tip coordinates to calculate the tooth root depth. . 10. The image of the saw blade size parameter described in claim 9 is wherein the method for measuring the radius of curvature of the tooth is searching for the coordinates of the edge of the saw blade image, and the coordinate arc of the tooth root is three points coordinate. ^, A, h are substituted into the following circular equations (1), (2), (3), and the three simultaneous equations can find three unknown coefficients C, D, E, and the obtained coefficients c, d, e Bring in the radius of curvature of the tooth (4): the equation of the circle male-anvil,,? The form of the red element is +Dy+E=〇, as in the formula (1), (2)' (3): (1) . (2) . (3) Cx, +Dy}+E = -( Xf + yf) Cx2 + Dy2 + E = -{xl + j Cx3 + Dy3 +E = -(3⁄4- + ^>32) x2V+Cx+Dy+E=0 =>(x+f )2+ (y+|)2=^+^EV 4 4 .......................{V 1 1 . As claimed in the patent application section 9 The image measuring method of the saw blade size parameter, wherein the saw blade gap angle measurement: 18 200944747 Obtain the coordinates (X,,),,),...,(u„) of the same long tangent line, and set y to X The linear regression line is ybv αχ The regression coefficient &, α is obtained by the least square method, so that Σ〇;, · -φ + αχ:)) 2 ί=ι is the smallest. The regression coefficients are solved as a=J^L— N- ^(x,· -x)2 /=1 b = y - αχ where 1 « ^=-Σ^· η /=ι 歹=-1^兄·η /=ι The clearance angle is θ = tan 1 ( 5). XI. Schema: as shown on page 19