201239351 五、發明說明：· 【發明所屬之技術領域】 本案係關於材料損傷檢測，尤指一種以瞬間熱傳導影 像之非破壞性檢測方法。 【先前技術】. 當一個較熱和一個較冷的物體接觸，能量便從較熱的 物體机向較冷的物冑’這種因溫度差異而引起的能量傳遞 為熱。熱傳遞主要有三種方式，分別是傳導、對流、輻射， 本案主要應用瞬間溫差之熱傳導方式作材料差異性檢測， 意即透過-短暫的時段中因傳導而致之溫差的變化、溫差 梯度來對材料内部的差異性如物質分布是否平均、是否 連續等特性進行檢測。由於溫度差而產生熱量從高溫區向 低咖區的轉移，兩物體間或同一物體的不同部位間，只要 存在溫差，就會發生熱量傳遞，直到各處溫度相同為止。 其中，熱傳導係數是一個物質的導熱性能，在同一物質内 :足高溫處傳到低溫處。本專利利用單位時間内，溫度變化 散失的熱能和内建的公式來計算熱傳導係數。 k = (Q/t) *L/(A*T) ⑴ 式中’ k是熱導率，q是熱傳導能量’歧瞬間熱傳導變化時 間，L是長度，a是接觸的面積，丁是利用瞬間加熱傳導特 I*生，在多層材料邊界層内會存在材料兩邊的溫度差，是發 生瞬間熱傳導過程的時間。 3 201239351 在現7產業中多層材料的研究已是一大領域，多層材 料是由二種或以上不同物質組成，組成後其機械性能及力 學性質可以優於原有組成材料的個別性質。其中一相為不 連繽、勁度大、強度高者稱為加強材，另一相為連續'低 勁度、低強度者稱為基材。多層材料之性質與組成材料之 個別性質、幾何形狀及相之分布情形有關，尤其以加強材 之體積比影響最大；加強材之分布決定多層材料系統是否 為均質性，若加強材分布不均勻，則多層材料便具異質性， 且有較高之可能由最脆弱處破壞，同時加強材之分佈方 向，亦造成多層材料具非等向性。一般高性能之多層材料， 其加強材為連續纖維，且在纖維方向之強度、勁度主要受 纖維所控制。至於基材，用以粘結、保護加強材、承載荷 重及加強材間局部應力之傳遞，並影響多層材料之機械性 質。多層材料對濕熱環境甚為敏感，以及材料内部若有損 壞’需要較複雜之非破壞檢驗方法偵測；至於製造方法， 則需要有較熟練之勞力及嚴格品質管制程序，但比起傳統 材料則可省去繁複的接頭設計及組立等。 此外’多層材料不論是在材料特性、強化結構'質量、 切削或磨耗等種種方面都可因其所需而做變更，而當進行 製程完成後可以利用光學非接觸式檢測多層材料的變形或 裂紋等情況’其應用的特性可直接量測應變場而非僅是位 移場’可減少許多繁瑣的運算及誤差的產生，在使用上更 為便捷。透過光學非接觸式之數位相機及數位影像處理系 統來觀測應變場’除了具有非接觸性、即時性、全場性的 4 201239351 優點外，更可以經由適當的擺設進而得到應變場的訊息， 由於其為利用光學干涉的原理進行量測，因此精確度相當 南。 材料在使用中的破壞損傷，或製程中缺陷與殘留應力 以及材料特性交互影響下，易造成整體結構性與功能性的 失效為儘早檢測出缺陷部位進行修復。傳統材料非破壞 性檢測方式有「液滲檢測」、「磁粒檢測」、「渴電流檢 則」 超音波檢測」、「射線檢測」等，其中「磁粒檢 測」與「渦電流檢測」方式僅適合於導磁性之金屬材料， 寸夕層複5材料無法應用。而「液滲檢測」僅適合於探測 表面的裂縫破壞，對於材料内部缺陷無法檢測。另外，「超 9波檢測」方法係以尚頻脈衡產生器產生電壓脈動，經由 同軸電境傳輸至換能II巾，換能^將電的脈波震I之超音 波而傳送入檢測物内，並接收來自表面、缺陷及底面等機 械震盪的回波，再轉換成脈動的電壓訊號。當檢測面粗糙 或曲面時，需要採用黏稠之耦合劑。超音波檢測時，探頭 選擇疋否適當關係檢測成敗，必須選用適宜的頻率及視檢 測物質需要製作各種不同形狀、大小及人工缺陷的校準規 k探頭與檢測面間之接觸面積大小，會改變檢測靈敏 度加上，只用能手動方式單點量測，無法全面性卽時量 、J耗時費力。「射線檢測」係利用高穿透能力的X或7 射線’在不傷害被照物體的本質狀況下而檢測出材料内部 非平面型的缺陷。然而，射線檢測方法所欲檢測材料内部 的4縫若與入射線方向垂直，很難被檢出。且X或γ射線201239351 V. INSTRUCTIONS: · [Technical field to which the invention pertains] This case relates to material damage detection, and more particularly to a non-destructive detection method for instantaneous heat conduction imaging. [Prior Art]. When a hotter one is in contact with a colder object, energy is transferred from a hotter object to a cooler object. This energy transfer due to temperature difference is heat. There are three main ways of heat transfer, namely conduction, convection and radiation. In this case, the thermal conduction mode of instantaneous temperature difference is mainly used for material difference detection, which means that the temperature difference caused by conduction and the temperature difference gradient are transmitted through a short period of time. The internal differences in the material, such as whether the material distribution is average, and whether it is continuous, are tested. The heat is transferred from the high temperature zone to the low coffee zone due to the temperature difference. If there is a temperature difference between the two objects or between different parts of the same object, heat transfer occurs until the temperatures are the same everywhere. Among them, the heat transfer coefficient is the thermal conductivity of a substance, which is transmitted to a low temperature in the same substance at a high temperature. This patent uses the heat energy lost during the unit time and the built-in formula to calculate the heat transfer coefficient. k = (Q/t) *L/(A*T) (1) where 'k is the thermal conductivity, q is the heat transfer energy', the instantaneous heat conduction change time, L is the length, a is the contact area, and the use is instantaneous. The heating conduction is special, and there is a temperature difference between the two sides of the material in the boundary layer of the multilayer material, which is the time during which the instantaneous heat conduction process takes place. 3 201239351 The research on multi-layer materials in the current 7 industry has been a large field. Multi-layer materials are composed of two or more different materials, and their mechanical properties and mechanical properties can be better than the individual properties of the original constituent materials. One of the phases is called non-continuous, high-strength, high-strength is called reinforcement, and the other phase is continuous 'low-stiffness, low-strength is called substrate. The properties of the multi-layer material are related to the individual properties, geometric shapes and distribution of the constituent materials, especially the volume ratio of the reinforcing material is the greatest; the distribution of the reinforcing material determines whether the multi-layer material system is homogeneous, and if the reinforcing material is unevenly distributed, The multilayer material is heterogeneous and has a higher probability of being destroyed by the most vulnerable, and the distribution direction of the reinforcing material also causes the multilayer material to be non-isotropic. In general, high-performance multilayer materials, the reinforcing material is continuous fiber, and the strength and stiffness in the fiber direction are mainly controlled by the fiber. As for the substrate, it is used to bond, protect the reinforcing material, load bearing weight and transfer of local stress between the reinforcing materials, and affect the mechanical properties of the multilayer material. Multi-layer materials are very sensitive to hot and humid environments, and if there is damage inside the material, 'more complex non-destructive inspection methods are needed. As for the manufacturing method, more skilled labor and strict quality control procedures are required, but compared with traditional materials. It can save complicated joint design and assembly. In addition, 'multilayer materials can be changed according to their needs in terms of material properties, reinforced structure' quality, cutting or abrasion, etc., and optical non-contact detection of deformation or crack of multilayer materials can be performed after the process is completed. In other cases, the characteristics of its application can directly measure the strain field rather than just the displacement field, which can reduce many cumbersome calculations and errors, and is more convenient to use. Observing the strain field through optical non-contact digital camera and digital image processing system. In addition to the advantages of non-contact, immediacy and full-field 4 201239351, it is possible to obtain the strain field information through proper placement. It is measured by the principle of optical interference, so the accuracy is quite south. Under the damage damage of the material during use, or the interaction between the defect and the residual stress and the material characteristics in the process, it is easy to cause the failure of the overall structural and functional aspects to repair the defect site as early as possible. Non-destructive detection methods for traditional materials include "liquid osmosis detection", "magnetic particle detection", "thirsty current detection", ultrasonic detection, and "ray detection". Among them, "magnetic particle detection" and "eddy current detection" methods are used. Only suitable for magnetically conductive metal materials, the inch layer layer 5 material can not be applied. The "liquid osmosis test" is only suitable for detecting crack damage on the surface, and it is impossible to detect internal defects of the material. In addition, the "Ultra 9-wave detection" method generates voltage pulsation by the frequency-frequency chronograph generator, and transmits it to the transducing II towel via the coaxial electric field, and transmits the ultrasonic wave of the pulse wave I to the detecting object. Internally, and receive echoes from mechanical oscillations such as surfaces, defects, and undersides, and then convert them into pulsating voltage signals. When detecting a rough surface or a curved surface, a viscous coupling agent is required. In the ultrasonic detection, the probe selection is not suitable for the detection success or failure. It is necessary to select the appropriate frequency and the detection material to be produced. The calibration area between the probes and the detection surface of various shapes, sizes and artificial defects will change the detection area. Sensitivity plus, only a single point measurement can be manually performed, it is impossible to comprehensively measure the amount of time, J is time-consuming and laborious. "Ray detection" uses high-permeability X or 7-rays to detect non-planar defects inside the material without damaging the nature of the object being illuminated. However, the four slits inside the material to be detected by the radiographic detection method are difficult to be detected if they are perpendicular to the incident direction. And X or gamma rays
S 201239351 夕/時會對人體造成輻射傷宏，且使檢制中从Μ ° 測出材料殘留放射 綠而烕為放射性物質。 友是，故’申請人有鑑於習知技術之缺失，發明了本 :「非破壞性檢測方法與系統」’用以改善上述習用技術手 段的缺失’並且可以應付上述習用技術所無法應付 内容。 容 内 明 發 發明之目的在於使用非破壞性的手段， 品内部是否有瑕疵，本發明所述非破壞性的手段是對待 物品施加一能量，並觀察待測物品的能量分布情形，從 判斷待測物品内部是否有瑕疵。更進一步而言，本發明 施加的肊罝疋一熱能。本發明的系統通常使用一加埶 備、一紅外線熱影像儀及一控制暨分析設備。溫度分佈 建；據此設計’將待測物品以加熱設備全面或局部加熱 並利用紅外線熱影像儀觀測待測物品内部熱源溫度分佈 形’再將該溫度分佈透過控制暨分析設備以等高線影像> 理或數值分析(Contour Map)及差異比對方法或是經由〗 影像明暗對比後’可判讀出材料之内部缺陷，更可找出乂 壞損傷位置及區域°譬如’當多層材料製程瑕㈣域之i 高線圖中的等高線密集處’即代表溫度變化梯度較大，^ 即代表瑕純大，通常是裂紋損傷變量大；反之，當多^ 材料製程瑕㈣域之等高線圖的等高線相當稀疏時，即j 示溫度梯度變化較小，代表瑕疵較小，㉟常是裂紋損傷餐 6 201239351 量小。且透過本發明的等高線圖所產生的形狀更可以輕易 的判斷待測物品内的瑕疵的形狀。 由此可見，本發明「非破壞性檢測方法與系統」，通常 是在一極短的時間内藉由熱傳導影像，在不傷害被待測物 體的本質狀況(物化特性)下而檢測出材料内部的缺陷，可 用於金屬、非金屬、複合材料及多層材料的檢測。且待测 材料之表面粗糙或曲而，处 飞曲面即表面狀態，對於紅外線熱影像 儀取像並不會有極重大的困難， π如 田難且取像範圍可以透過拍攝 範圍而加以改變，故本發 现尽發明具有全面性即時量測，對於 物品不會產生損壞，可快速檢測較廣區域、清楚定位缺 二：：、探測淺層内部缺陷、亦可檢測金屬、非金屬材料、 或複合材料’可以改善上述習用手段之缺失。 【實施方式】 以下針對本案之非破壞性 ^ ^ ^錢檢財法的各實施例進行福 述，4參考附圖，但實際之 —a # Α ι夂所彳木仃的方法並不必須 兀王符.所描述的内容，熟、 钗π考田月匕在不脫離本案 之實際精神及_的情況下，做出種種變化及修改。 本發月疋對-待測物品施加—能量 待測物品上的分佈壯β t， I檢心此里在 ^佈^，來判斷此相物 以及瑕疵的型態。更進 瑕疵、 一待測物本發明是施加一熱能在 待測物…並利 散發出的紅外線輕射* ^ 俄來測量待測物品所 了在π時稗: 外線輕射能的分佈即表示 時待測物品内部的熱源分佈情形，並基於物趙表S 201239351 On the evening of the day, it will cause radiation damage to the human body, and the material will be emitted from Μ ° during the inspection. Friends, therefore, the applicant has invented the "non-destructive testing method and system" to improve the lack of the above-mentioned conventional technical means in view of the lack of the prior art, and can cope with the above-mentioned conventional technology. The purpose of the invention is to use non-destructive means to determine whether there is any flaw inside the product. The non-destructive means of the invention is to apply an energy to the article and observe the energy distribution of the article to be tested. Test if there is any flaw inside the item. Still further, the present invention applies a thermal energy. The system of the present invention typically uses a heating device, an infrared thermal imager, and a control and analysis device. The temperature distribution is built; according to the design, the article to be tested is heated by the heating device in whole or in part and the infrared heat imager is used to observe the internal heat source temperature distribution shape of the object to be tested, and then the temperature distribution is transmitted through the control and analysis device to the contour image > Contour Map and the difference comparison method or the internal defects of the material can be judged by comparing the light and dark contrast of the image, and the location and area of the damaged damage can be found. For example, when the multilayer material process (four) domain The i-line in the high-line diagram is dense, which means that the temperature gradient is large, ^ means that the 瑕 is large, usually the crack damage variable is large; on the contrary, when the contour line of the multi-material process 四(4) domain is quite sparse When, that is, j shows that the temperature gradient changes little, which means that the 瑕疵 is small, and 35 is often the crack damage meal 6 201239351. Further, the shape of the flaw in the article to be tested can be easily judged by the shape produced by the contour map of the present invention. It can be seen that the "non-destructive detection method and system" of the present invention generally detects the inside of the material by thermally transmitting the image in a very short period of time without damaging the essential condition (physical property) of the object to be tested. Defects for the detection of metals, non-metals, composites and multilayers. Moreover, the surface of the material to be tested is rough or curved, and the surface of the flying surface, that is, the surface state, does not have a great difficulty for the infrared thermal imager to take images, and the π image is difficult and the image capturing range can be changed through the shooting range. Therefore, this invention has a comprehensive and accurate measurement method, which can not damage the articles, and can quickly detect a wide area and clearly locate the defects:: detecting shallow internal defects, and detecting metal, non-metal materials, or Composite materials' can improve the lack of the above-mentioned conventional means. [Embodiment] The following is a description of the embodiments of the non-destructive ^ ^ ^ money check method of the present invention, 4 with reference to the drawings, but the actual method of - a # Α 夂 夂 彳 并不 并不 does not have to be 兀Wang Fu. The content described, familiar, 钗π Kao Tianyue made various changes and modifications without departing from the actual spirit and _ of the case. This month's 疋 疋 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Further, the present invention is to apply a thermal energy to the object to be tested and to emit the infrared light rays*^ to measure the object to be tested at π: the distribution of the external light energy is The distribution of heat sources inside the object to be tested, and based on the object
7 S 201239351 面/皿度與輕射能成函數關係的原理，透過一控制暨分析設 備’將來自紅外線攝影機的資料予以自動計算並透過螢幕 顯不觀測與計算結果，如溫度數值與等高線圖。 紅外線是指電磁波譜中介於可見光與微波間之波段， 由於波長比紅外光長，故稱為紅外線。紅外線波長範圍很 廣，依波長大小又可分為近紅外線、中紅外線、遠紅外線。 自然界任何物體只要溫度高於絕對零度卜273)就會產生 電礙波’ p遺著其表面溫度之高低，會輻射出不$能量之紅 卜線其度愈局，所輻射的總能量愈大》紅外線熱影像 儀即疋以感應物體熱輻射能而產生影像，不同於傳統的檢 測法有操作費時以及檢測區域小等缺點。本專利採用紅外 線…、心像儀檢測技術來量測多層材料的内部溫度分布情 形0 "月參閱® 1 ’為本發明的非破壞性檢測系统的示意圖。 其中可見-加熱設備卜内容納了 —待測物品4,待測物品 4内4則可能具有瑕疵4’，而在加熱設備^的附近則設置 红外線熱影像儀2，具有一熱影像範圍2，，用以拍攝待 測物品4於加熱設備】内受熱之後的熱能分佈的狀況。加 熱設備1可以是自水浴也 ，由，合烤相、蒸氣室、微波爐中 選擇一種，亦即，凡是可以使—物體溫度上升者均可以是 本發明的加熱設備。而紅外線熱影像儀2更與—分析設備 電連接、.工外線熱衫像儀2將拍攝到的原始影像數據傳 送給分析設備3,並透過其來建立以等高線圖或數值分析 8 201239351 來顯不的溫度分佈。操作步嫌普土 韦作’騾百先是將待測件瞬間局部加 熱，利用紅外線熱影像儀監測玄麻 ㈣4夕層#才斗内部熱源溫度分 情形，再將該溫度分佈以等离结 寺间線影像處理或數值分析 (C〇ntour Map)及差異比對方 找®破壞才貝傷位置及區 域。譬如檢測一多層材料，其内部的損傷、瑕疲，會造成 熱傳遞（傳導）時的阻礙，使得熱傳導分佈不均勻，進而造 成該處的溫度有著較為劇烈的差異，此即產生了溫度： 度，透過紅外線熱影像儀即可觀測到此溫度梯度。此外， 由於待測物品4本身製好後就具有一定的溫度，因此除了 以加熱的方式來觀察内部埶 ，，、、你/皿度刀佈之外，亦可以透過 冷卻的方式來觀察其埶量嵛 如仏 、里政失的狀況，如果是無瑕疵的待 測物品4則應該是整體的、&、木人，、 體均勻的被冷卻’若其内部有瑕疵則 透過紅外線熱影像亦可觀窣 J祝祭到值度分佈不均勻的情形。本 發明為了便於理解起自禮I、，l ^ < , 文見僅以加熱手段作為說明，但對於熟 悉本技術領域的人而言， 透過冷郃來觀察紅外線熱影像分 佈疋顯而易見的。 此外’熱流場分析技術是以數位影像相關係數法技 -亩：了可以分析待測物品之溫度分佈重建，#裂縫周圍 ^破㈣傷位置及區域。紅外線㈣像儀還可以記錄 你从 電腦進竹影像分析、粒子識別和粒子位 H相關算法得到瞬時熱流場分析。 在D兄月等鬲線影像處理或數值分析(Contour Map)及差 、對方法’找出破壞損傷位置及區域之前，先定義裂紋 9 201239351 #傷變量的定義。以下以多層材料為例。 請參閱圖以至圖2d，其中，圖h為無瑕庇的待測物品 的縱剖視圖、_為有瑕疫的待測物品的縱剖視圖、圖^ 為無瑕疵的待測物品的熱源分佈示意圖、而圖2“有瑕疵 的待測物品的熱源分佈示意圖…圖2c與圖礙將多層 材料之瑕疵區域所造成的溫度梯度分佈以等高線圖 (Contour Map)標# ’並計算多層材料瑕庇區域之裂紋損傷 變量’做為非破壞性檢測方法。…丨代表第一層多層材 料崖2代表第一層多層材料，…，I代表第n層多層材料。 利用、’工外線熱影像儀配合瞬間加熱傳導特性’觀測多層材 料損傷的現象’多層材料裂紋損傷變量定義為下（Du) : 式中’ Η!代表最低層多層材料損傷變量開裂層厚度，&代 表次低層多層材料損傷變量開裂層厚度’而到^代表最高 層多層材料損傷變量開裂層厚度。Dn為兩相鄰裂紋間距。 以下將陸續介紹如何利用紅外線熱影像儀觀測瑕疵區域之 品域IV像D十算其影像大小並配合瞬間溫度差異觀測多層 材料瞬間熱傳導影像來分析損傷的現象。 明參閱圖3a至圖3e。圖3a為一有瑕疵待測物品受檢 測：不思圖’待測物品4具有一瑕疵4,，並被涵蓋在熱影 像蚝圍2内。圖3b為熱影像儀觀測瑕疵區域示意圖，其中 可見在熱影像範圍2,内的影像是紅外線熱影像，而在該範 圍外的則疋一般可見光影像，則完全看不出有何變化。圖 201239351 3c為瑕疵區域的紅外線熱影像示意圖，即是將熱影像範圍 2’涵蓋住瑕疵區域而拍攝出的紅外線熱影像；而圖3(j則是 可見光影像示意圖，而在拍攝區域内有無瑕疵不會在可見 光影像上看到任何變化。如此可見圖3b是以圖3d為背景， 將圖3c的示意圖置於該瑕疵所在的相對位置，以表現出肉 眼觀測與實際熱影像的相對關係。圖3e為待測物品内部熱 源分佈的等溫線（C〇nt〇ur Map)示意圖。此是將同一溫度的 位置以一封閉曲線連接起來，一如氣象圖中的等壓線的概 念。因此透過圖3e的各等溫線的距離即可經過計算而得到 此瑕疵區域的損傷變量。不過僅憑圖3c與圖3e即可判斷 此待測物品在此區域内有材料分佈不均勻、不連續的狀況 存在。不過，由於為了縱軸座標的顯示符合一般座標系的 直觀，因此，圖3e的等溫線圖剛好與圖3c呈上下顛倒， 雖然如此，但這並不影響上述的關於瑕疵區域損傷變量的 測量與計算。 此外圖3 c的熱影像儀觀測瑕疲區域示意圖是一灰階 影像，而實際上熱影像則都是彩色色階、通常是&光譜形 式來表現，故請參見附件2，為圖3c的原始圖式，由附件 2可見其影像是彩色色階的紅外線熱影像，經由彩色效果 可以更為具體的感受到溫度的分佈。同理，圖％的原始圖 式請見附件卜其中可見在熱影像範圍2，（請配合圖叫内 的影像是彩色色階的紅外線熱影像，經由彩色效果可以更 為具體的感受到溫度的分佈。在附件i中央的淺紅色區域 對照右側的表尺可知此區域溫度是最高的，並逐漸向外遞 11 201239351 月 > 見附件3，此為可見光圖式，亦即圖3d的原始圖 f、附件3中的紅色框則代表圖3b的熱影像範圍y，由於 '見光來表現，所以看不出來有任何的變化。又，圖 是-單純的黑白圖式’但原始圖式是一彩色等溫線圖， 請參考附件4。 昭請參閱圖4a至圖4e，這些圖式為前述圖3的各圖的對 …、、’且其中，圖4a為無瑕疵待測物品受檢測的示意圖。圖 外為熱影像儀觀測無瑕疵區域示意圖’其中可見熱影像範 圍内毫無異狀。圖4c為無瑕疵區域的紅外線熱影像示意 圖而圖4d則是可見光影像示意圖。如此可見圖α是以 圖牝為背景’將圖4c的示意圖置於該瑕疵所在的相對位 置’以表現出肉眼觀測與實際熱影像的相對關係。圖46為 待測物品内部熱源分佈的等高線(c〇nt〇urMap)示意圖。此 是將同-溫度的位置以-封閉曲線連接起來，一如氣象圖 中的等壓線的概念。因此圖4e的等高線實際上即是一等溫 線’透過各等溫線的距離即可經過計算而得#此瑕疲區域 的損傷變量。由此可見，在無瑕疵的形況之下，圖讣中不 會見到顯示出不同溫度的熱影像，而圖补不會顯示出等高 線圖，此即表示溫度的分佈極為平均幾+沒有#度變化， 故據此可確定在此區域内沒有材料分佈不均勻、或不連續 的狀況存在。不過’由於為了縱軸座標的顯示符合一般座 標系的直觀，因此，圖4e的等溫線圖剛好與圖4c呈上下 顛倒，雖然如此，但這並不影響上述的關於瑕疵區域損傷 變量的測量與計算。請參閱附件5 '附件6、與附件7，分 12 2012393517 S 201239351 The principle of the relationship between surface/dish and light-emission can be calculated automatically by a control and analysis device. The data from the infrared camera is automatically calculated and displayed through the screen, such as temperature values and contour maps. Infrared is a wavelength band between visible light and microwave in the electromagnetic spectrum. Since the wavelength is longer than infrared light, it is called infrared. The infrared wavelength range is wide, and can be classified into near-infrared, mid-infrared, and far-infrared depending on the wavelength. Any object in nature, as long as the temperature is higher than the absolute zero degree 273), it will produce an electric wave. The surface temperature of the surface will be radiated, and the red line of the energy will be radiated, and the total energy of the radiation will be greater. 》Infrared thermal imager is the image that produces the image by sensing the heat radiation energy of the object. It is different from the traditional detection method, which has the disadvantages of time-consuming operation and small detection area. This patent uses infrared ray..., cardiograph detection technology to measure the internal temperature distribution of multilayer materials. 0 "Monthsee® 1 ' is a schematic diagram of the non-destructive detection system of the present invention. It can be seen that the heating device contains the object to be tested 4, and the object 4 in the object to be tested 4 may have 瑕疵4', and in the vicinity of the heating device ^, the infrared thermal imager 2 is provided, which has a thermal image range of 2, The condition for collecting the thermal energy distribution after the object 4 to be tested is heated in the heating device. The heating device 1 may be a self-water bath, and may be selected from a grilling phase, a steam chamber, and a microwave oven, that is, any one that can raise the temperature of the object may be the heating device of the present invention. The infrared thermal imager 2 is further electrically connected to the analysis device, and the external image camera 2 transmits the captured raw image data to the analysis device 3, and through which the contour map or numerical analysis is established 8 201239351 No temperature distribution. The operation step is suspected to be the first step of the general test of the parts to be tested, and the infrared heat imager is used to monitor the internal heat source temperature of the Xing Ma (4) 4 层 layer #, and then the temperature distribution is equal to the temple. Line image processing or numerical analysis (C〇ntour Map) and the difference between the other side to find the damage to the position and area of the shell. For example, when detecting a multi-layer material, the internal damage and fatigue will cause obstacles in heat transfer (conduction), resulting in uneven distribution of heat conduction, which in turn causes a sharp difference in temperature, which results in temperature: This temperature gradient can be observed through an infrared thermal imager. In addition, since the object to be tested 4 itself has a certain temperature after being prepared, in addition to observing the internal flaws by heating, and/or the knife cloth, it can also be observed by cooling. If the quantity is such as 仏, the situation of the government is lost, if it is the innocent item to be tested 4, it should be the whole, &, the wooden person, the body is evenly cooled 'if there is a flaw inside, the infrared thermal image is also transmitted. Obviously, I wished that the value of the value distribution was uneven. The present invention has been described with reference to heating means only for ease of understanding, but it is apparent to those skilled in the art that the infrared thermal image distribution is observed by cold heading. In addition, the 'thermal flow field analysis technique is based on the digital image correlation coefficient method-mu: the temperature distribution of the object to be tested can be analyzed and reconstructed, and the location and area of the wound around the crack are broken. The infrared (four) imager can also record the instantaneous thermal flow field analysis from the computer image analysis, particle identification and particle position H correlation algorithm. Define the crack 9 201239351 #伤变量's definition before the D-brother and so on line image processing or numerical analysis (Contour Map) and the difference and the method to find the damage damage location and area. The following is an example of a multilayer material. Please refer to the figure and FIG. 2d, wherein FIG. h is a longitudinal sectional view of the object to be tested without a shelter, _ is a longitudinal sectional view of the object to be tested with plague, and FIG. 2 is a schematic diagram of the heat source distribution of the object to be tested without flaws. Fig. 2 Schematic diagram of the heat source distribution of the object to be tested (Fig. 2c and Fig. 2c) The temperature gradient distribution caused by the area of the multilayer material is plotted by the contour map (Contour Map) and the crack of the multi-layer material is calculated. The damage variable 'as a non-destructive test method....丨 represents the first layer of multi-layer material cliff 2 represents the first layer of multi-layer material, ..., I represents the n-th layer of multi-layer material. Use, 'external line thermal imager with instantaneous heating conduction Characteristic 'observation of multi-layer material damage phenomenon' Multi-layer material crack damage variable is defined as lower (Du): where Η! represents the lowest layer multilayer material damage variable crack layer thickness, & represents the sub-low layer multilayer material damage variable crack layer thickness And ^ represents the maximum layer multilayer material damage variable cracking layer thickness. Dn is the spacing between two adjacent cracks. The following will introduce how to use infrared thermal imager The product area IV of the 瑕疵 area is imaged by D, and the instantaneous heat conduction image of the multilayer material is observed to analyze the damage phenomenon. For details, refer to Fig. 3a to Fig. 3e. Fig. 3a shows a defective object to be tested: Do not think that the item to be tested 4 has a 瑕疵4, and is covered in the thermal image circle 2. Figure 3b is a schematic view of the 瑕疵 area of the thermal imager, which can be seen in the thermal image range 2, the image is infrared heat The image, while outside the range, the general visible light image, there is no change at all. Figure 201239351 3c is a schematic diagram of the infrared thermal image of the 瑕疵 region, that is, the thermal image range 2' covers the 瑕疵 area and is taken Infrared thermal image; and Figure 3 (j is a schematic view of visible light, and there is no change in the visible image in the shooting area. So it can be seen that Figure 3b is in the background of Figure 3d, the schematic of Figure 3c Placed in the relative position of the raft to show the relative relationship between the visual observation and the actual thermal image. Figure 3e is the isotherm of the internal heat source distribution of the object to be tested (C〇nt Ur map). This is to connect the positions of the same temperature with a closed curve, just like the concept of isobars in the weather map. Therefore, the distance from each isotherm in Fig. 3e can be calculated and obtained. The damage variable of the area. However, it can be judged by Fig. 3c and Fig. 3e that the material to be tested has uneven and discontinuous material distribution in this area. However, since the display of the coordinates of the vertical axis conforms to the general coordinate system Intuitive, therefore, the isotherm diagram of Figure 3e is just upside down with Figure 3c. However, this does not affect the above-mentioned measurement and calculation of the damage variable in the 瑕疵 region. In addition, the thermal imager of Figure 3 c observes fatigue. The regional schematic is a grayscale image, but in fact the thermal image is color gradation, usually & spectroscopy, so please refer to Appendix 2, which is the original image of Figure 3c. The image is visible in Annex 2. The infrared thermal image of the color gradation can more specifically feel the temperature distribution through the color effect. For the same reason, the original pattern of the figure % can be found in the attached image. It can be seen in the thermal image range 2, (please match the image in the picture is the infrared thermal image of the color level, and the temperature effect can be more specific through the color effect. Distribution. In the light red area in the center of the attachment i, the meter on the right side is the highest, and the temperature is the highest. It is gradually outward. 11 201239351 Months> See Annex 3, which is the visible pattern, which is the original image of Figure 3d. The red box in Annex 3 represents the thermal image range y of Figure 3b. Since it is represented by the light, there is no change. However, the figure is a simple black and white pattern, but the original pattern is one. For the color isotherm diagram, please refer to Appendix 4. Please refer to Fig. 4a to Fig. 4e, which are the pairs of the above diagrams of Fig. 3, and wherein, in Fig. 4a, the innocent object to be tested is tested. Schematic. Outside the picture is a schematic diagram of the flawless area of the thermal imager. It shows no abnormalities in the thermal image range. Figure 4c is a schematic diagram of the infrared thermal image of the flawless area and Figure 4d is a schematic diagram of the visible light image. Taking the map as the background 'put the schematic of Fig. 4c at the relative position of the raft' to show the relative relationship between the naked eye observation and the actual thermal image. Fig. 46 is the contour of the heat source distribution of the object to be tested (c〇nt〇urMap Schematic. This is to connect the position of the same-temperature with a closed curve, just like the concept of isobar in the weather map. Therefore, the contour of Figure 4e is actually an isotherm 'through each isotherm The distance can be calculated to obtain the damage variable of the fatigue zone. It can be seen that under the innocent condition, no thermal image showing different temperatures will be seen in the image, and the patch will not show up. Contour map, this means that the temperature distribution is extremely average + no change of # degree, so it can be determined that there is no uneven or discontinuous material distribution in this area. However, due to the display of coordinates for the vertical axis The general coordinate system is intuitive. Therefore, the isotherm diagram of Figure 4e is just upside down with Figure 4c. However, this does not affect the above measurement and calculation of the damage variable in the 瑕疵 region. See Annex 5 'Attachment 6, and Annex 7, on 12 201239351
別是圖4b、圖4C、同/L _ 4C圖4d的原始彩色 對於圖4b的埶I僮 在附件5中相The original color of Figure 4b, Figure 4C, and /L _ 4C Figure 4d. For the 埶I child of Figure 4b, in Annex 5
…、〜像範圍2，的區域內的& A A 小，對昭旁邊$矣 /色色階變化很 J釕”，、牙遠的表尺可見在埶影傻笳 化很…不同於附…‘，、、“象祀圍2區域内的溫度變 、附件1所呈現的狀態。附 顯示熱影像範圍2，P主觖人因 什〇則疋早獨 固2 U配合圖4b)的熱影像圖式。 件7，此為可見光圖式，紅 / 7, , ^ θ 1V 巴I 代表圖仆的熱影像範圍 2 ’由於疋以可見光來表現， 由圖3b、圖4b、附件3 “ 出來有任何的變化’ 下，不… 附件7可知’在可見光的情形之 “則物品内部是否有瑕疵，都是幾乎無法辨識 的。又’圖4e是一單純的黑白圖式，但原始圖式是一彩色 等溫線圖，請參考附件8， 以說是沒有。 此了見-度的梯度變化可 通常在一個影像平面上，物件的移動可藉由影像序列 中不同影像粒子灰階分佈變化觀測得到的。而空間中的不 同影像粒子間流場分佈情形相對於影像上稱為光流場，反 映了影像上每-影像粒子點灰階變化趨勢，由於本發明中 影像粒子點的灰階變化實際上就代表該位置的溫度變化， 故本發明即依據此灰階變化趨勢，來推導出溫度變化的趨 :。本發明是透過瞬間加熱裝備、紅外線熱影像儀及 孤度刀佈重建，據此設計，將待測件瞬間局部加熱，利用 紅外線熱影像儀監測多層材料内部熱源溫度分佈情形，再 將該溫度分部以等高線影像處理或數值分析（c〇nt〇ur Map) 及差異比對方法，找出破壞損傷位置及區域。其中，熱流 場反映了，5V像上母一影像粒子點灰階變化趨勢，即該影像 5 13 201239351 粒子點所在位置的溫度變化 以將帶有灰階的像素點在影 度場，當作溫度變化速度場 可推導熱流場方程如下： ，並透過單位時間的觀察，可 像平面上運動而產生的瞬時速 ，也是一種真實熱流場估計， 假設E(x，y，t)為影像點㈣在㈣t(觀測的起始時 的灰階值。設⑽時刻（自該起始時間經過—單位時間 該影像粒子點移動至影像點(x+dx，y+dy)，其灰階值為 E(x+dX，y+dy，t+dt)。由於對應同一影像粒子點，以E(x，y，^ E(x+dx，y+dy，t+dt)代表熱流場約束方㈣㈣泰勒展開式計 算，並令dt->〇，可得到Exu+Eyv+Et = 〇，其中，& =肛/叔 Ey = dE/dy Et = dE/dt u = dx/dt v = _卜求解熱應力流場 約束方程得到u，v。當攝影機於固定不動情形下，觀測影 像粒子點移動變化，即可分離熱流場及背景影像問題。對 於背景影像’理想情況下，其熱流應當為Q，只有熱流場 才有熱流。因此，不用熱流約束方程求出U，V，只要求出 影像亮度梯度方向的速率，即可求出sqrt(u*u+v*v)。由熱 流約束方程可求得梯度方向的光流速率為V 1 abs(Et/Sqrt(Ex*Ex+Ey*Ey))，設定一個閾值 T，得到 v(x y) >τ’影像點（x，y)為熱流場；v(x，y) $ τ，影像點（xy)為 背景。 請參閱圖5a至圖5h各圖。其中，圖5a為紅外線熱影 像儀觀測起始時間⑴之瑕疵區域示意圖、圖5b為红外線熱 影像儀觀測所經過一單位時間（dt)後（t+dt)之瑕疵區域視示 14 201239351 意圖，此二圖是以灰階顯示’而其中心影像顯示高溫且白 熱化。圖5c與圖5d分別為圖5a與圖5b之瑕疵區域的等 高線（Contour Map)示意圖’圖5c與圖5d的原始彩色等.、田 線圖請分別參閱附件9與附件10、圖5e為觀測影像粒子 點移動變化趨勢圖、圖5f為圖5a與圖5b熱流場影像分佈 圖初始化狀態示意圖，圖5f是灰階的紅外線熱影像圖，但 由於僅是初始化’故而溫度梯度並不明顯且圖5f上所劃分 出的三個溫域的溫差實際上十分接近、圖5g為瑕疵區域的 熱量不均勻分佈之熱流場分佈示意圖、以及圖5h為瑕疵區 域局。卩不均勻分佈之熱流場分佈圖，其原始的彩色等溫線 圖請看附件11。由此可見，透過兩個時點所觀測到的紅外 線影像粒子的變化，可以顯示出熱流場，而熱流場反映了 影像上每一影像粒子點灰階變化趨勢，可看成是帶有灰階 的像素點在影像平面上運動而產生的瞬時速度場，也是一 種真實熱流場估計。不過，由於為了縱軸座標的顯示符合 二般座標系、的直觀，因此，圖5d的等溫線圖剛好 分別與圖5a及圖5b呈上下顛倒，同理，圖5h亦是圖& 的上下顛包j，雖然如& ’但這並不影響上㉛的關於瑕疵區 域損傷變量的測量與計算。 I . ’T、上所述，本發明透過施加一熱能在一待測物品，並 :用、、工外線熱影像儀，來測量待測物品所散發出的 yo 6J. 2¾½. . 此，而此紅外線輻射能的分佈即表示了在高溫時待測 °»内部的熱源分佈情形，並基於物體表 成函數關係的原理，透過一控制暨分析設備，絲二 g. 15 201239351 線攝影機的資料予以自動計算並透過螢幕顯示觀測 結果，如溫度數值與等高線圖。此外，並更進一步的透過 觀測時間的經過’進一步的分析兩個時點所觀測到的紅外 ^影像粒子的變化’可以顯示出熱流場，❿熱流場反映了 影像上每-影像粒子點灰階變化趨勢，可看成是帶有灰階 的像素點在影像平面上運動而產生的瞬時速度場，也是一 種真實熱流場估計。簡而言之，本發明的透過加熱的方式， 來檢測待測物品是否有瑕疲，其原理在於若待測物品内的 驗佈是均句的’則其被加熱後，其熱能的分佈也應該 疋均勻的’因&，若待測物品内有瑕疵’則材料分佈即不 勻’也就導致了熱能分部不均勻。相對的，亦可以將加熱 的方式改為冷卻’若冷卻後發現有熱量散失不均勻的^ 態’亦可以確認待測物品内有瑕寐。由此觀之，本 於檢測一物品內县：^： 士 ^ 否有瑕疵，提供了-種近乎無害且應用 範圍極廣的非破壞性檢測方法與系統，對於物品檢測技術 的發展具有重大的貢獻。 、】技術 實施例： 發明提供一種非破壞性檢測系 設備，用以加埶一待 ^ … ， 、]物°口，以及一紅外線熱影像儀，用 觀測該待測物品所發出的紅外線熱影像。 2. 如實施例 水浴、油浴、烤箱 3. 如實施例 熱影像。 所述的系統，其中該加熱設備可以是自 '热氣室、微波爐中選擇一種。 1所述的系統，其中該檢測結果是紅外線 16 201239351 4.如實施例1所述的系統，其中該檢測結果是等高線 圖原理繪製的等溫線圖（Contour Map)。 5 .如實施例1所述的系統，更包括一分析設備，將一 個或一個以上之來自該紅外線熱影像儀的資料予以分析， 以形成一檢測結果。 6.本發明提供一種非破壞性檢測方法，包括下列步 驟：提供一待測物品；加熱該待測物品；以及擷取來自該 待測物品的熱分佈。 7·如實施例6所述的方法’其中該熱分佈是以紅外線 熱影像呈現。 ★ ^ 8·如實施例7所述的方法，其中該紅外線熱影像是以 等高線技術所繪製的等溫線圖。 9·—種非破壞性檢測方法，包括下列步驟：提供一待 測物品；加熱該待測物品；以及擷取來自該待測物品的兩 個時點的熱分佈。 10.如實施例9所述的方法，其中更包括—步驟：將 各時點的各該熱分佈處理為一熱流場影像。 上述實施例僅係為了方便說明而舉例，雖遭熟悉本技 藝之人士任施匠思而為諸般修飾，然皆不脫如附申請專利 範圍所欲保護者。 【圖式簡單說明】 圖1，為本發明的非破壞性檢測系統的示意圖； 圖2 a，為無瑕庇的待測物品的縱剖視圖. 17 g. 201239351 圖2b，為有瑕疲的待測物品的縱剖視圖； 圖2c ’為無瑕疵的待測物品的熱源分佈示意圖； 圖2d為有瑕疵的待測物品的熱源分佈示意圖； 圖3a，為有瑕疵待測物品受檢測的示意圖； 圖3b，為熱影像儀觀測瑕疵區域示意圖； 圖3c，為瑕疵區域的紅外線熱影像示意圖； 圖3d，為可見光影像示意圖； 圖3e ’為待測物品内部熱源分佈的等高線示意圖； 圖4a，為無瑕疵待測物品受檢測的示意圖； 圖4b，為熱影像儀觀測無瑕疵區域示意圖； 圖4c，為無瑕疵區域的紅外線熱影像示意圖； 圖4d，為可見光影像示意圖； 圖4e，為待測物品内部熱源分佈的等高線示意圖； 圖5a，為紅外線熱影像儀觀測起始時間⑴之瑕疵區域 不意圖， 圖5b，為紅外線熱影像儀觀測所經過一單位時間（价) 後（t+dt)之瑕疲區域視示意圖； 圖5c與圖5d’分別為圖5a與圖5b之瑕疵區域的等高 線（Contour Map)示意圖； 圖5e，為觀測影像粒子點移動變化趨勢圖； 圖5f，為圖5a與圖讣熱流場影像分佈圖初始化狀態 示意圖； 18 201239351 圖5g，為瑕疵區域的熱量不均勻分佈之熱流場分佈示 意圖；以及 圖5h，為瑕庇區域局部不均勻分佈之熱流場分佈圖。 附件1，為圖3b的原始的以彩色色階影像顯示之紅外 線熱影像，在熱影像範圍2,外的則是可見光影像； 附件2，為圖3c的原始的以彩色色階影像顯示之紅外 線熱影像； 附件3 ’為圖3d的原始圖式； 附件4,為圖3e的原始圖式之彩色等溫線圖； 附件5，為圖4b的原始的以彩色色階影像顯示之紅外 線熱影冑，在熱影像範® 2,外的則是可見光影像； 附件6,為圖心的原始的以彩色色階影像顯示之紅外 線熱影像； 附件7 ’為圖4d的原始圖式，紅色框則代表圖a的 熱影像範圍2,； 附件8，為圖4e的原始的彩色等溫線圖； 附件9 ’為圖5c的原始彩色等溫線圖 附件1〇 ’為圖5d的原始彩色等溫線圖；以及 附件11，為圖5h的原始彩色等溫線圖。 【主要元件符號說明】 1 :加熱設備 2 :紅外線熱影像儀..., ~ like the range 2, the area within the & AA is small, the side of the 矣 矣 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 色 牙 牙 牙 牙 牙 牙 牙 牙 牙 牙 牙 牙 牙, "The temperature in the area around the area 2, and the state shown in Annex 1. Attached to the thermal image range 2, P main 因 〇 独 独 独 独 独 独 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 。 。 Item 7, this is the visible light pattern, red / 7, , ^ θ 1V Bar I represents the thermal image range of the servant 2 'Because 疋 is expressed in visible light, there are any changes from Figure 3b, Figure 4b, Annex 3 'Under, no... Attachment 7 knows that 'in the case of visible light', there is a flaw in the inside of the article, which is almost indistinguishable. Also, Fig. 4e is a simple black and white pattern, but the original pattern is a color isotherm, please refer to Appendix 8, to say no. This gradient of see-degree can usually be on an image plane, and the movement of the object can be observed by the change of the gray scale distribution of different image particles in the image sequence. The flow field distribution between different image particles in space is called the optical flow field on the image, which reflects the gray-scale change trend of each image particle point on the image. Because the gray-scale change of the image particle point in the present invention is actually Representing the temperature change at this position, the present invention derives the trend of temperature change based on this gray-scale change trend. The invention is reconstructed by an instant heating equipment, an infrared thermal imager and a solitary knife cloth. According to the design, the workpiece to be tested is locally heated instantaneously, and the temperature distribution of the heat source inside the multilayer material is monitored by an infrared thermal imager, and then the temperature is divided. The contour map processing or numerical analysis (c〇nt〇ur Map) and the difference comparison method are used to find out the location and area of the damage. Among them, the heat flow field reflects the change of the gray level of the 5V image on the image of the parent image, that is, the temperature change of the position of the particle point of the image 5 13 201239351 to treat the pixel with gray scale in the shadow field as the temperature The variable velocity field can be used to derive the heat flow field equation as follows: and through the observation of unit time, the instantaneous velocity generated by the motion on the plane is also a true heat flow field estimation, assuming E(x, y, t) is the image point (4) (d) t (the gray level value at the beginning of the observation. Set (10) time (from the start time - unit time the image particle point moves to the image point (x + dx, y + dy), the gray level value is E ( x+dX, y+dy, t+dt). Because of the same image particle point, E(x, y, ^ E(x+dx, y+dy, t+dt) represents the heat flow field constraint (4) (4) Taylor expansion Calculate, and let dt-> 〇, you can get Exu+Eyv+Et = 〇, where & = anal / uncle Ey = dE / dy Et = dE / dt u = dx / dt v = _ The stress flow field constraint equation obtains u, v. When the camera is stationary, the image particle point movement changes can be observed, and the heat flow field and background image problem can be separated. For the background image, ideally, the heat flow should be Q, and only the heat flow field has heat flow. Therefore, U, V can be obtained without the heat flow constraint equation. As long as the velocity of the image brightness gradient direction is obtained, the sqrt can be obtained. u*u+v*v). The heat flow constraint equation can be used to obtain the light velocity rate in the gradient direction V 1 abs(Et/Sqrt(Ex*Ex+Ey*Ey)), and set a threshold T to obtain v(xy >τ' image point (x, y) is the heat flow field; v(x, y) $ τ, image point (xy) is the background. Please refer to Fig. 5a to Fig. 5h, wherein Fig. 5a is infrared heat The schematic diagram of the area after the observation time of the imager (1), and Fig. 5b shows the intention of the area after the unit time (dt) of the infrared thermal imager (t+dt) 14 201239351, the two figures are gray scale The display shows 'the center image shows high temperature and white heat. Figure 5c and Figure 5d are the contour maps of the contours of Figure 5a and Figure 5b respectively. The original color of Figure 5c and Figure 5d. Refer to Annex 9 and Annex 10, Figure 5e for the observed image particle point movement change trend, Figure 5f for Figure 5a and Figure 5b Schematic diagram of the initialization state of the flow field image distribution map, Fig. 5f is the infrared thermal image of the gray scale, but since it is only initialized, the temperature gradient is not obvious and the temperature difference of the three temperature domains divided on Fig. 5f is actually very close. Fig. 5g is a schematic diagram of the heat flow field distribution of the uneven distribution of heat in the 瑕疵 region, and Fig. 5h is the 瑕疵 regional bureau. The heat flow field distribution map of the 卩 uneven distribution, the original color isotherm diagram of the seeing is shown in Annex 11. It can be seen that the change of the infrared image particles observed at two time points can show the heat flow field, and the heat flow field reflects the gray scale change trend of each image particle point on the image, which can be regarded as a gray scale. The instantaneous velocity field generated by the movement of pixels on the image plane is also a true heat flow field estimate. However, since the display of the coordinates of the vertical axis conforms to the intuition of the two coordinate systems, the isotherm diagram of FIG. 5d is just upside down with FIG. 5a and FIG. 5b respectively. Similarly, FIG. 5h is also the diagram of FIG. Up and down the package j, although as & 'but this does not affect the measurement and calculation of the upper 31 damage variable. I. 'T, as described above, the present invention measures the yo 6J. 23⁄41⁄2. emitted by the object to be tested by applying a heat energy to an object to be tested, and using an external line thermal imager. The distribution of the infrared radiant energy indicates the distribution of the heat source inside the °» at high temperature, and based on the principle of the functional relationship between the objects, through a control and analysis device, the information of the wire camera is given by the g. 15 201239351 line camera. Automatically calculate and display observations such as temperature values and contour maps through the screen. In addition, and further through the observation time, 'further analysis of the changes in the infrared image particles observed at two time points' can show the heat flow field, and the heat flow field reflects the gray-scale change of each image particle point on the image. The trend can be seen as the instantaneous velocity field generated by the movement of pixels with grayscale on the image plane, and is also a true heat flow field estimation. In short, the method of detecting heat in the present invention detects whether the article to be tested is fatigued. The principle is that if the inspection in the article to be tested is a uniform sentence, then the heat energy distribution is also Should be evenly 'causes & if there is 瑕疵 in the item to be tested, then the material distribution is uneven', which leads to uneven heat energy. In contrast, it is also possible to change the heating mode to cooling 'If the heat is found to be uneven, the heat is lost. From this point of view, in the inspection of an article in the county: ^: 士^ No 瑕疵, provides a non-destructive detection method and system that is almost harmless and has a wide range of applications, which is of great significance for the development of article detection technology. contribution. Technical Example: The invention provides a non-destructive detection system device for adding a device, a port, and an infrared thermal imager for observing the infrared thermal image emitted by the object to be tested. . 2. As in the example water bath, oil bath, oven 3. As in the example Thermal image. The system wherein the heating device can be selected from the group consisting of a 'hot air chamber, a microwave oven. The system of claim 1, wherein the detection result is infrared ray 16 201239351 4. The system of embodiment 1, wherein the detection result is a contour map drawn by a contour plot principle (Contour Map). 5. The system of embodiment 1 further comprising an analysis device for analyzing one or more data from the infrared thermal imager to form a test result. 6. The present invention provides a non-destructive detecting method comprising the steps of: providing an item to be tested; heating the item to be tested; and extracting a heat distribution from the item to be tested. 7. The method of embodiment 6 wherein the heat distribution is presented as an infrared thermal image. The method of embodiment 7, wherein the infrared thermal image is an isotherm diagram drawn by a contour technique. 9. A non-destructive testing method comprising the steps of: providing an item to be tested; heating the item to be tested; and extracting a heat distribution from the two points of time of the item to be tested. 10. The method of embodiment 9, further comprising the step of: processing each of the heat distributions at each time point into a heat flow field image. The above-described embodiments are merely examples for the convenience of the description, and those skilled in the art will be able to modify them without departing from the scope of the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic view of a non-destructive detection system of the present invention; Figure 2a is a longitudinal cross-sectional view of an article to be tested without a shield. 17 g. 201239351 Figure 2b, for a fatigue test Longitudinal section view of the article; Figure 2c' is a schematic diagram of the heat source distribution of the innocent object to be tested; Figure 2d is a schematic diagram of the heat source distribution of the object to be tested; Figure 3a is a schematic view of the object to be tested being tested; Figure 3c is a schematic diagram of the infrared thermal image of the 瑕疵 region; Figure 3d is a schematic view of the visible light image; Figure 3e' is a schematic diagram of the contour distribution of the heat source inside the object to be tested; Figure 4a is a flawless view Figure 4b is a schematic diagram of the flawless area of the object to be tested; Figure 4c is a schematic diagram of the infrared thermal image of the flawless area; Figure 4d is a schematic view of the visible light image; Figure 4e is the inside of the object to be tested Schematic diagram of the contour distribution of the heat source; Figure 5a, the infrared thermal imager observation start time (1) is not intended, Figure 5b, is the infrared thermal image Figure 5c and Figure 5d' are schematic diagrams of the Contour Map of the area between Figure 5a and Figure 5b, respectively; Figure 5e, Figure 5e, is a schematic view of the area of the fatigue after a unit of time (price) (t + dt); Observing image particle point movement change trend diagram; Figure 5f is a schematic diagram of the initial state of the heat flow field image distribution map of Fig. 5a and Fig. 18 201239351 Fig. 5g is a schematic diagram of the heat flow field distribution of the heat uneven distribution in the 瑕疵 region; and Fig. 5h, A heat flow field distribution map that is unevenly distributed locally in the shelter area. Attachment 1 is the original infrared thermal image displayed in color gradation image of Figure 3b. In the thermal image range 2, the visible image is visible; Annex 2 is the original infrared color image displayed in color gradation image of Figure 3c. Thermal image; Annex 3 'is the original image of Figure 3d; Annex 4 is the color isotherm of the original image of Figure 3e; Annex 5 is the original infrared thermal image of the color gradation image of Figure 4b胄, in the thermal image Fan® 2, the outside is the visible light image; Annex 6 is the original infrared thermal image displayed in the color gradation image of the heart; Annex 7 'is the original image of Figure 4d, the red frame is Represents the thermal image range 2 of Figure a; Annex 8 is the original color isotherm diagram of Figure 4e; Annex 9 'is the original color isotherm of Figure 5c. Attachment 1〇' is the original color isothermal of Figure 5d. Line graph; and Annex 11, is the original color isotherm plot of Figure 5h. [Main component symbol description] 1 : Heating equipment 2 : Infrared thermal imager
S 19 201239351 2’ ：熱影像範圍 3 ·分析設備 4 :待測物品 4’ ：瑕疵 20S 19 201239351 2' : Thermal image range 3 · Analytical equipment 4 : Items to be tested 4' : 瑕疵 20