TW202147251A - Method for estimating state of liquid level, computer program product, and computer readable recording medium - Google Patents
Method for estimating state of liquid level, computer program product, and computer readable recording medium Download PDFInfo
- Publication number
- TW202147251A TW202147251A TW109119212A TW109119212A TW202147251A TW 202147251 A TW202147251 A TW 202147251A TW 109119212 A TW109119212 A TW 109119212A TW 109119212 A TW109119212 A TW 109119212A TW 202147251 A TW202147251 A TW 202147251A
- Authority
- TW
- Taiwan
- Prior art keywords
- liquid level
- region
- positioning
- area
- interest
- Prior art date
Links
Images
Landscapes
- Image Analysis (AREA)
- Measurement Of Levels Of Liquids Or Fluent Solid Materials (AREA)
Abstract
Description
本發明係關於一種液面狀態評估技術，特別是關於一種基於煉鋼模液面的影像串流建立的液面狀態估測方法、電腦程式產品及電腦可讀取紀錄媒體。 The present invention relates to a liquid level state evaluation technology, in particular to a liquid level state estimation method, a computer program product and a computer-readable recording medium based on an image stream of the liquid level of a steelmaking mold.
在鋼鐵冶煉領域中，模液面的液面擾動(如液面滾盪)是造成許多煉鋼缺陷的成因，習知煉鋼缺陷偵測技術多是使用渦電流(eddy current)或熱相儀(thermal phase analyzer)進行量測。 In the field of iron and steel smelting, the liquid level disturbance of the mold liquid level (such as liquid level tumbling) is the cause of many steelmaking defects. The conventional steelmaking defect detection techniques mostly use eddy current or thermal phase meter. (thermal phase analyzer) for measurement.
以紅外線熱像儀(或稱紅外線熱成像儀)為例，係對模液面進行攝影產生熱影像串流資料，再以人工觀測熱影像中的液面滾盪的情況。然而，以人眼觀察影像的結果除了無法量化，也容易產生標準不一及作業耗時等問題。 Taking an infrared thermal imager (or infrared thermal imager) as an example, the liquid surface of the mold is photographed to generate thermal image streaming data, and then the liquid surface tumbling in the thermal image is manually observed. However, in addition to being unable to quantify the results of viewing images with the human eye, problems such as inconsistent standards and time-consuming operations are prone to occur.
實務上，此類滾盪造成的缺陷，例如捲渣缺陷，多會發生在邊蓋區，如以習知渦電流方法偵測液面擾動，由於渦電流只能量測空間中的單點固定位置之狀態，無法針對容易發生捲渣缺陷之邊蓋區持續進行量測。 In practice, defects caused by such tumbling, such as slag entrapment defects, mostly occur in the side cover area. For example, the conventional eddy current method is used to detect liquid level disturbance, because eddy current can only measure a single point fixed in space. The status of the position cannot be continuously measured for the edge cover area that is prone to slag defects.
有鑑於此，有必要提供一種有別以往的技術方案，以解決習知技術所存在的問題。 In view of this, it is necessary to provide a different technical solution to solve the problems existing in the prior art.
本發明之一目的在於提供一種液面狀態估測方法，基於熱影像中的感興趣區定位在易產生捲渣缺陷的區域附近，對相鄰影像的感興趣區位置比對差異，以利有效降低煉鋼缺陷。 One of the objectives of the present invention is to provide a liquid level state estimation method, based on which the region of interest in the thermal image is positioned near the area prone to slag defects, and the differences in the positions of the regions of interest in adjacent images are compared to facilitate effective Reduce steelmaking defects.
本發明之次一目的在於提供一種電腦程式產品，基於熱影像中的感興趣區定位在易產生捲渣缺陷的區域附近，對相鄰影像的感興趣區位置比對差異，以利有效降低煉鋼缺陷。 Another object of the present invention is to provide a computer program product, based on which the region of interest in the thermal image is positioned near the area prone to slag defects, and the differences in the position of the region of interest in adjacent images are compared, so as to effectively reduce the thermal Steel defects.
本發明之再一目的在於提供一種內儲程式之電腦可讀取紀錄媒體，基於熱影像中的感興趣區定位在易產生捲渣缺陷的區域附近，對相鄰影像的感興趣區位置比對差異，以利有效降低捲渣缺陷。 Another object of the present invention is to provide a computer-readable recording medium with a stored program. Based on the location of the region of interest in the thermal image near the area prone to slag defects, the location of the region of interest in adjacent images is compared. difference, in order to effectively reduce the slag defects.
為達上述之目的，本發明的一方面提供一種液面狀態估測方法，被配置成由一處理器執行被儲存在一記憶體中的指令，該方法包含步驟：獲取一熱影像串流資料，該熱影像串流資料包含朝向一加熱中的液體表面以離散時間方式攝取的多幅熱影像，各該熱影像包括一液面區域及在該液面區域周圍的至少一容器邊緣；將該些熱影像中的每個逐一設定為一目標影像，對該目標影像進行：在該液面區域周圍搜尋該至少一容器邊緣作為至少一定位基線；將該液面區域內的一感興趣區定位在該定位基線附近；將該感興趣區映對到該目標影像的至少一相鄰影像的區域設定為至少一比對熱區；及估測該比對熱區與該感興趣區之間的一差異程度；及依據該差異程度產生一狀態指標。 In order to achieve the above object, one aspect of the present invention provides a liquid level state estimation method, which is configured to be executed by a processor to execute instructions stored in a memory, the method comprising the steps of: acquiring a thermal image stream data , the thermal image stream data includes a plurality of thermal images taken in discrete time mode towards a heating liquid surface, each of the thermal images includes a liquid surface area and at least one container edge around the liquid surface area; the Each of these thermal images is set as a target image one by one, and the target image is: searched for the at least one container edge around the liquid level area as at least a positioning baseline; locate a region of interest in the liquid level area near the positioning baseline; setting the region of the region of interest mapped to at least one adjacent image of the target image as at least one comparison hot region; and estimating the distance between the comparison hot region and the region of interest a difference degree; and generating a state index according to the difference degree.
在本發明之一實施例中，該些熱影像的狀態指標經過均化處理而產生一均化狀態曲線，判斷該均化狀態曲線是否大於一警示閾值，若判斷為是，輸出一警示訊號，若判斷為否，輸出一紀錄訊號。 In one embodiment of the present invention, the state indicators of the thermal images are subjected to a homogenization process to generate a homogenized state curve, and it is determined whether the homogenized state curve is greater than a warning threshold. If the judgment is no, output a record signal.
在本發明之一實施例中，在該液面區域周圍搜尋該至少一容器邊緣作為該至少一定位基線的步驟包括：在該液面區域周圍偵測在至少一直線軌跡上的至少一邊緣線段；及依據在同一直線軌跡上的所有邊緣線段產生該至少一定位基線。 In one embodiment of the present invention, the step of searching for the at least one container edge around the liquid level area as the at least one positioning baseline includes: detecting at least one edge line segment on at least one linear trajectory around the liquid level area; and generating the at least one positioning baseline according to all the edge segments on the same straight line.
在本發明之一實施例中，該至少一定位基線依據該至少一邊緣線段及一雜訊濾波演算法產生。 In an embodiment of the present invention, the at least one positioning baseline is generated according to the at least one edge line segment and a noise filtering algorithm.
在本發明之一實施例中，該定位基線的數量為二，該二定位基線相互交會在一點。 In an embodiment of the present invention, the number of the positioning baselines is two, and the two positioning baselines intersect each other at one point.
在本發明之一實施例中，將該感興趣區映對到該目標影像的在前影像及/或在後影像的區域設定為該比對熱區。 In an embodiment of the present invention, a region where the region of interest is mapped to the previous image and/or the subsequent image of the target image is set as the comparison hot region.
在本發明之一實施例中，該感興趣區的至少一邊線與該目標影像的至少一邊線平行，該感興趣區的二相鄰邊線之間的一交點被設定為一定位點，從該定位點到該至少一定位基線中的一個的一定位距離不大於形成該定位點的二相鄰邊線的長度總和。 In an embodiment of the present invention, at least one side line of the region of interest is parallel to at least one side line of the target image, and an intersection between two adjacent side lines of the region of interest is set as an anchor point. A positioning distance from the positioning point to one of the at least one positioning baseline is not greater than the sum of the lengths of two adjacent edge lines forming the positioning point.
在本發明之一實施例中，該狀態指標為依據該差異程度產生的一差異機率。 In an embodiment of the present invention, the state indicator is a difference probability generated according to the difference degree.
為達上述之目的，本發明的另一方面提供一種電腦程式產品，當電腦載入該電腦程式並執行後，該電腦能夠執行如上所述之液面狀態估測方法。 In order to achieve the above object, another aspect of the present invention provides a computer program product. After the computer program is loaded and executed, the computer can execute the liquid level state estimation method as described above.
為達上述之目的，本發明的另一方面提供一種電腦可讀取紀錄媒體，該電腦可讀取紀錄媒體內儲程式，當電腦載入該程式並執行後，該電腦能夠完成如上所述之液面狀態估測方法。 In order to achieve the above-mentioned purpose, another aspect of the present invention provides a computer-readable recording medium, the computer can read a program stored in the recording medium, and after the computer loads the program and executes it, the computer can complete the above-mentioned Liquid level state estimation method.
本發明的液面狀態估測方法、電腦程式產品及電腦可讀取紀錄媒體，通過將該液面區域內的感興趣區定位在該定位基線附近；將該感興趣區映對到該目標影像的至少一相鄰影像的區域設定為至少一比對熱區；及估測該比對熱區與該感興趣區之間的差異程度，依據該差異程度產生該目標影像的狀態指標，可用於煉鋼製程中追蹤邊蓋位置調整該量測區域，利用邊蓋位置定位該量測區域離邊蓋之距離，可以有效偵測較易發生問題的邊蓋週邊情況，以利有效降低煉鋼缺陷。 The liquid level state estimation method, computer program product and computer-readable recording medium of the present invention locate the region of interest in the liquid level region near the positioning baseline; map the region of interest to the target image The area of at least one adjacent image is set as at least one comparison hot area; and the degree of difference between the comparison hot area and the area of interest is estimated, and the state indicator of the target image is generated according to the degree of difference, which can be used for In the steelmaking process, the position of the side cover is tracked to adjust the measurement area. Using the position of the side cover to locate the distance between the measurement area and the side cover can effectively detect the surrounding conditions of the side cover that are prone to problems, so as to effectively reduce steelmaking defects. .
P1:取像步驟 P1: Image acquisition steps
P2:測邊步驟 P2: Steps for edge measurement
P3:定位步驟 P3: Positioning step
P4:映對步驟 P4: Mapping step
P5:估測步驟 P5: Estimation step
P6:示警步驟 P6: Warning steps
P:區間 P: interval
R:區間 R: interval
R1:感興趣區 R1: Region of Interest
D1:定位距離 D1: positioning distance
D2:定位距離 D2: Positioning distance
L1:定位基線 L1: locate the baseline
L2:定位基線 L2: Positioning the baseline
M:目標影像 M: target image
A:定位點 A: Anchor point
S1:相似度 S 1 : Similarity
S2:相似度 S 2 : Similarity
S3:相似度 S 3 : Similarity
S4:相似度 S 4 : Similarity
S5:相似度 S 5 : Similarity
S6:相似度 S 6 : Similarity
t:時間軸 t: Timeline
第1圖：本發明實施例之液面狀態估測方法的流程示意圖。 Fig. 1 is a schematic flowchart of a liquid level state estimation method according to an embodiment of the present invention.
第2圖：本發明實施例之液面狀態估測方法的感興趣區的定位示意圖。 Fig. 2 is a schematic diagram of the positioning of the region of interest of the liquid level state estimation method according to the embodiment of the present invention.
第3圖：本發明實施例之液面狀態估測方法針對第一區域影像進行差異數值分析的示意圖。 FIG. 3 is a schematic diagram of the difference numerical analysis performed on the image of the first region by the liquid level state estimation method according to the embodiment of the present invention.
第4圖：本發明實施例之液面狀態估測方法在不同畫幅的第二區域內搜索與該第一畫幅的該第一區域中影像相似的區域的示意圖。 FIG. 4 is a schematic diagram of searching for an area similar to an image in the first area of the first frame in the second area of the different frame by the liquid level state estimation method according to the embodiment of the present invention.
第5圖：本發明實施例之之液面狀態估測方法的相鄰畫幅的感興趣區差異估算的示意圖。 FIG. 5 is a schematic diagram of the difference estimation of the region of interest of adjacent frames in the liquid level state estimation method according to the embodiment of the present invention.
第6圖：本發明實施例之液面狀態估測方法經過連續的比對及連續計算處理後得到的連續時間滾盪數據的示意圖。 Fig. 6 is a schematic diagram of continuous time rollover data obtained after continuous comparison and continuous calculation processing by the liquid level state estimation method according to the embodiment of the present invention.
第7圖：本發明實施例之液面狀態估測方法經過數值分析處理後轉換成狀態指標的示意圖。 Fig. 7 is a schematic diagram of converting the liquid level state estimation method into a state index after numerical analysis and processing according to an embodiment of the present invention.
為了讓本發明之上述及其他目的、特徵、優點能更明顯易懂，下文將特舉本發明較佳實施例，並配合所附圖式，作詳細說明如下。再者，本發明所提到的方向用語，例如上、下、頂、底、前、後、左、右、內、外、側面、周圍、中央、水平、橫向、垂直、縱向、軸向、徑向、最上層或最下層等，僅是參考附加圖式的方向。因此，使用的方向用語是用以說明及理解本發明，而非用以限制本發明。 In order to make the above-mentioned and other objects, features and advantages of the present invention more clearly understood, the preferred embodiments of the present invention will be exemplified below and described in detail in conjunction with the accompanying drawings. Furthermore, the directional terms mentioned in the present invention, such as up, down, top, bottom, front, rear, left, right, inner, outer, side, surrounding, center, horizontal, lateral, vertical, longitudinal, axial, Radial, uppermost or lowermost, etc., are only directions with reference to the attached drawings. Therefore, the directional terms used are for describing and understanding the present invention, not for limiting the present invention.
本發明的一方面提供一種液面狀態估測方法，可由一處理器(processor)執行被儲存在一記憶體(memory)中的指令，該記憶體耦接該處理器，該記憶體還可儲存相關資料(例如待比對資料及比對結果等)，例如該處理器可以被配置成一資料處理裝置、平台或伺服器等，該記憶體可以被配置成一資料儲存媒體或資料庫等，該記憶體與該處理器可以被整合為一體，在此並不設限。應被理解的是，該液面狀態可指為液體表面的靜態或動態特徵，例如：液面波動或擾動(fluctuation)等狀態，諸如加熱中的液體的液面滾盪狀態，或者，在液體中通入氣體而在液面產生氣泡或氣泡破滅狀態等，但不以此為限。如第1圖所示，該方法包含：一取像步驟P1、一測邊步驟P2、一定位步驟P3、一映對步驟P4及一估測步驟P5。 An aspect of the present invention provides a method for estimating a liquid level. A processor can execute instructions stored in a memory, the memory is coupled to the processor, and the memory can also store instructions Relevant data (such as data to be compared and comparison results, etc.), for example, the processor can be configured as a data processing device, platform or server, etc., the memory can be configured as a data storage medium or database, etc., the memory The body and the processor may be integrated into one, which is not limited herein. It should be understood that the liquid level state may refer to static or dynamic characteristics of the liquid surface, such as: liquid level fluctuation or disturbance (fluctuation) state, such as the liquid level tumbling state of the liquid under heating, or, in the liquid. Gas is introduced into the liquid surface to generate bubbles or bubble bursting state, etc., but not limited to this. As shown in FIG. 1 , the method includes: an image capturing step P1 , an edge measuring step P2 , a positioning step P3 , a mapping step P4 and an estimation step P5 .
該取像步驟P1，可獲取一熱影像串流資料，該熱影像串流資料包含朝向一加熱中的液體表面以離散時間方式攝取的多幅熱影像，各該熱影像包括一液面區域及在該液面區域周圍的至少一容器邊緣。 In the imaging step P1, a thermal image stream data can be acquired, and the thermal image stream data includes a plurality of thermal images captured in a discrete-time manner toward a heating liquid surface, and each thermal image includes a liquid surface area and At least one container rim around the liquid level area.
該測邊步驟P2，可將該些熱影像中的每個逐一設定為一目標影像，對該目標影像進行：在該液面區域周圍搜尋該至少一容器邊緣作為至少一定位基線。 In the edge measuring step P2, each of the thermal images can be set as a target image one by one, and the target image is searched for the at least one container edge around the liquid level area as at least one positioning baseline.
該定位步驟P3，可將該液面區域內的一感興趣區(region of interest，ROI)定位在該定位基線附近。 In the positioning step P3, a region of interest (ROI) in the liquid surface area may be positioned near the positioning baseline.
該映對步驟P4，可將該感興趣區映對到該目標影像的至少一相鄰影像的區域設定為至少一比對熱區(hot zone for comparison)。 In the mapping step P4, a region where the region of interest is mapped to at least one adjacent image of the target image can be set as at least one hot zone for comparison.
該估測步驟P5，可估測該比對熱區與該感興趣區之間的一差異程度；及依據該差異程度產生一狀態指標。 In the estimation step P5, a degree of difference between the comparison hot area and the area of interest may be estimated; and a state indicator may be generated according to the degree of difference.
可選地，在一實施例中，如第1圖所示，該方法還可包含一示警步驟P6，可將該些熱影像的狀態指標經過均化處理(例如採用window averaging algorithm)而產生一均化狀態曲線，用於判斷該均化狀態曲線是否大於一警示閾值，若判斷為是，輸出一警示訊號，若判斷為否，輸出一紀錄訊號。 Optionally, in an embodiment, as shown in FIG. 1 , the method may further include a warning step P6 , in which the state indicators of the thermal images may be subjected to an averaging process (eg, using a window averaging algorithm) to generate a The averaged state curve is used for judging whether the averaged state curve is greater than a warning threshold. If the judgment is yes, a warning signal is output, and if the judgment is no, a record signal is output.
以下舉例說明上述液面狀態估測方法實施例的實施態樣，在此僅以估測煉鋼用的模液面的擾動狀態為例，例如：可用於估測煉鋼用的模液通入氣體(如氬氣)後在液面產生氣泡或氣泡破滅狀態等，但不以此為限，惟不以此為限，該液面狀態估測方法實施例也可用於估測其餘液體(例如水)的液面狀態。 The following examples illustrate the implementation of the above-mentioned method for estimating the liquid level. Here, only the disturbance state of the mold liquid level for steelmaking is taken as an example. For example, it can be used to estimate the introduction of mold liquid for steelmaking. After the gas (such as argon), bubbles are generated on the liquid surface or the bubbles are burst, but it is not limited to this. water) of the liquid level.
舉例而言，該處理器可從該記憶體獲取該熱影像串流資料，該熱影像串流資料可線上或離線儲存於該記憶體，例如該熱影像串流資料可由紅外線熱像儀朝向一加熱中的液體表面(例如煉鋼用的模液面)以離散時間(discrete time)方式攝取多幅熱影像所形成，例如該些熱影像是在時間軸被間歇性攝取的，該些熱影像可以是彩色熱影像，也可以是彩色熱影像經過灰階處理後的灰階熱影像，其顏色深淺對應不同的溫度數值。 For example, the processor can obtain the thermal image stream data from the memory, and the thermal image stream data can be stored in the memory online or offline, for example, the thermal image stream data can be directed by an infrared thermal imager toward a The heating liquid surface (such as the mold liquid surface for steelmaking) is formed by capturing multiple thermal images in discrete time. For example, these thermal images are intermittently captured on the time axis. It can be a color thermal image, or a gray-scale thermal image after the color thermal image is processed by grayscale, and the color depth corresponds to different temperature values.
如第2圖所示，該處理器可將該些熱影像中的每個逐一設定為一目標影像M，以灰階熱影像為例，各該熱影像包括一液面區域(見圖中淺色長條區域，如模液面所在區域等)及在該液面區域周圍的至少一容器邊緣(見圖中深色背景區域，如邊蓋等)。 As shown in FIG. 2 , the processor can set each of the thermal images as a target image M one by one. Taking gray-scale thermal images as an example, each thermal image includes a liquid surface area (the shallow area in the figure). color strip area, such as the area where the mold liquid level is located, etc.) and at least one container edge around the liquid level area (see the dark background area in the figure, such as the side cover, etc.).
如第2圖所示，該處理器可對該目標影像M進行：在該液面區域周圍搜尋該至少一容器邊緣(如L1)作為至少一定位基線，但不以此為限；例如該定位基線的數量也可以為二個(如L1、L2)，該二定位基線L1、L2可相互交會在一點，以利進行定位，作為後續進行該液面狀態估測方法的依據。 As shown in FIG. 2 , the processor may perform on the target image M: search for the at least one container edge (such as L1 ) around the liquid level area as at least one positioning baseline, but not limited thereto; for example, the positioning The number of baselines may also be two (eg L1 and L2 ). The two positioning baselines L1 and L2 may intersect with each other at one point to facilitate positioning and serve as the basis for the subsequent liquid level state estimation method.
可選地，在一實施例中，如第2圖所示，在該液面區域周圍搜尋該至少一容器邊緣作為該至少一定位基線(如L1、L2)的步驟可包括：在該液面區域周圍偵測在至少一直線軌跡上的至少一邊緣線段，例如可採用邊緣偵測法(諸如Hough轉換法或RANSCA隨機式尋找演算法等)來偵測該邊緣線段；及依據在同一直線軌跡上的所有邊緣線段產生該至少一定位基線，例如可將在一第一直線軌跡上的所有邊緣線段連接並延長，以形成該定位基線L1，並將在一第二直線軌跡上的所有邊緣線段連接並延長，以形成該定位基線L2。藉此，可以利用該至少一定位基線作為在邊線(如邊蓋)附近評估液面狀態(如滾盪狀態等)的定位依據，可以協助判定邊蓋附近的模液面是否發生滾盪現象，以利有效降低煉鋼缺陷。 Optionally, in an embodiment, as shown in FIG. 2, the step of searching for the at least one container edge around the liquid level area as the at least one positioning baseline (eg L1, L2) may include: at the liquid level Detecting at least one edge line segment on at least one linear trajectory around the area, for example, an edge detection method (such as Hough transform method or RANSCA random search algorithm, etc.) can be used to detect the edge segment; and based on the same linear trajectory The at least one positioning baseline can be generated by all edge line segments of Extend to form the positioning baseline L2. In this way, the at least one positioning baseline can be used as the positioning basis for evaluating the liquid level state (such as tumbling state, etc.) near the side line (such as the side cover), which can help determine whether the mold liquid surface near the side cover has a tumbling phenomenon. In order to effectively reduce steelmaking defects.
可選地，在一實施例中，該至少一定位基線還可依據該至少一邊緣線段及一雜訊濾波演算法產生(諸如Kalman濾波等遞歸濾波演算法)。藉此，如果遇到畫面模糊(例如衍生自鑄模過程出現震動及其他晃動)，則可利用該雜訊濾波演算法持續追蹤該容器邊緣(如邊蓋)，可將突然劇烈變化的容器邊緣偵測結果去除，以減少錯誤偵測結果影響後續估測結果。 Optionally, in one embodiment, the at least one positioning baseline can also be generated according to the at least one edge line segment and a noise filtering algorithm (such as a recursive filtering algorithm such as Kalman filtering). In this way, if the image is blurred (such as vibration and other shaking caused by the molding process), the noise filtering algorithm can be used to continuously track the edge of the container (such as the side cover), and the edge of the container that changes suddenly and violently can be detected. The measurement results are removed to reduce the impact of false detection results on subsequent estimation results.
後續，該處理器可對該目標影像進行：將該液面區域內的一感興趣區定位在該定位基線(如可為單一或多條定位基線)附近，作為後續進行該液面狀態估測方法的依據。 Subsequently, the processor may perform on the target image: a region of interest in the liquid level area is positioned near the positioning baseline (for example, there may be single or multiple positioning baselines), as the subsequent estimation of the liquid level state basis of the method.
可選地，在一實施例中，如第2圖所示，該感興趣區R1的至少一邊線與該目標影像M的至少一邊線平行，該感興趣區R1的二相鄰邊線之間的一交點被設定為一定位點A，從該定位點A到該至少一定位基線中的一個(如L1)的一定位距離(如A與L1的中垂線距離)D1不大於形成該定位點A的二相鄰邊線的長度總和，用於定位該感興趣區R1。具體地，例如該感興趣區可選擇易因液面滾盪產生煉鋼缺陷的位置，如：邊蓋旁；可選地，也可以使用多個感興趣區佈滿該液面區域進行液面全域偵測，但不以此為限。 Optionally, in one embodiment, as shown in FIG. 2 , at least one side line of the region of interest R1 is parallel to at least one side line of the target image M, and the space between two adjacent side lines of the region of interest R1 is parallel. An intersection is set as a positioning point A, and a positioning distance (such as the mid-perpendicular distance between A and L1) D1 from the positioning point A to one of the at least one positioning baseline (such as L1) is not greater than the positioning point A formed The sum of the lengths of the two adjacent edges is used to locate the region of interest R1. Specifically, for example, the region of interest can select a location where steelmaking defects are likely to be generated due to liquid level tumbling, such as: beside the side cover; Global detection, but not limited to this.
但不以此為限，該感興趣區R1的定位規則也可以依據實際情況加以改變，例如該定位點A到該至少一定位基線中的另一個(如L2)的一定位距離(如A與L2的中垂線距離)D2不大於形成該定位點A的二相鄰 邊線的長度總和。 But not limited to this, the positioning rule of the region of interest R1 can also be changed according to the actual situation, for example, a positioning distance (such as A and A and L2) from the positioning point A to the other (such as L2) of the at least one positioning baseline. The distance from the mid-perpendicular line of L2) D2 is not greater than the two adjacent to form the positioning point A The sum of the lengths of the edges.
後續，該處理器可對該目標影像進行：將該感興趣區映對到該目標影像的至少一相鄰影像的區域設定為至少一比對熱區，例如：如第3圖所示，可基於一時間軸t，將該感興趣區(如圖所示之熱影像中的矩形框)映對到該目標影像的在前影像及/或在後影像的區域設定為該比對熱區(如圖所示之熱影像中的矩形框)，其中該矩形框僅用於表示該比對熱區與感興趣區是在不同熱影像中的位置相同的區域。 Subsequently, the processor may perform on the target image: setting the region of interest mapped to at least one adjacent image of the target image as at least one comparison hot area, for example, as shown in FIG. Based on a time axis t, the region where the region of interest (the rectangular frame in the thermal image as shown) is mapped to the previous image and/or the subsequent image of the target image is set as the comparison hot region ( The rectangular frame in the thermal image as shown in the figure), wherein the rectangular frame is only used to indicate that the comparison thermal region and the region of interest are regions in the same location in different thermal images.
後續，該處理器可對該目標影像進行：估測該比對熱區與該感興趣區之間的一差異程度，依據該差異程度產生一狀態指標，例如依據該差異程度產生的一差異機率。 Subsequently, the processor may perform on the target image: estimate a degree of difference between the comparison hot area and the area of interest, and generate a status indicator according to the degree of difference, such as a probability of difference generated according to the degree of difference .
舉例說明如下，以該熱影像串流資料包含該目標影像與其前或後相鄰影像構成的三個畫幅為例進行說明，惟不以此為限。 An example is described below, and the thermal image stream data includes three frames formed by the target image and its preceding or succeeding adjacent images as an example, but it is not limited thereto.
舉例而言，以熱液(如煉鋼用模液面)的滾盪狀態為例，由於熱影像中的像素顏色深淺對應不同的溫度數值，當熱影像中的液面區域(如模液面)發生滾盪現象時，整個液面區域中的像素顏色深淺變化代表局部溫度產生變化，因此，若連續的熱影像的像素之間變動越大，代表發生滾盪現象的機率越高。其中，考量估算的範圍與其精度有關，為了避免範圍過大影響估測可信度，可用該感興趣區作為一估算範圍。其中，一位置的液面滾盪機率可以如下式1表示： For example, taking the tumbling state of the hot liquid (such as the mold liquid level for steelmaking) as an example, since the color depth of the pixels in the thermal image corresponds to different temperature values, when the liquid level area in the thermal image (such as the mold liquid level) ) When the tumbling phenomenon occurs, the change in the color depth of the pixels in the entire liquid surface area represents the local temperature change. Therefore, the greater the variation between pixels in consecutive thermal images, the higher the probability of tumbling. Wherein, the estimated range is considered to be related to its accuracy. In order to avoid the influence of too large range on the estimation reliability, the region of interest can be used as an estimated range. Among them, the probability of liquid surface tumbling at a position can be expressed by the following formula 1:
此外，如果要量測一個區域(如該感興趣區)內的滾盪機率，可以將該區域內的滾盪機率依上式1積分得到，如下式2所示： In addition, if you want to measure the roll probability in a region (such as the region of interest), the roll probability in the region can be obtained by integrating the above formula 1, as shown in the following formula 2:
此外，要達到比較一特定區域的差異數值，可以對時間微 分。上述的微分可以利用高通濾波達到微分的效果。在該影像資料中的前後數個時間區間內的該第一畫幅、該第二畫幅及該第三畫幅中，針對一第一區域(如該感興趣區中的像素方格)內影像和前後畫幅作比較，以得到該第一區域內影像的差異數值。 In addition, to compare the difference values for a particular area, you can point. The above-mentioned differentiation can use high-pass filtering to achieve the effect of differentiation. In the first frame, the second frame and the third frame in several time intervals before and after the image data, for the image in a first area (such as the pixel square in the region of interest) and the front and rear The frames are compared to obtain the difference value of the images in the first area.
舉例來說，請參照第3圖所示，針對該第一畫幅、該第二畫幅及該第三畫幅中的該第一區域內影像進行差異數值分析。上述的差異數值分析可以通過下式表示： For example, referring to FIG. 3 , a difference numerical analysis is performed on the images in the first area in the first frame, the second frame and the third frame. The above-mentioned difference numerical analysis can be expressed by the following formula:
或 or
根據上式的差異數值分析，可以將上式2代表的一滾盪程度改寫由下式3表示： According to the difference numerical analysis of the above formula, the rolling degree represented by the above formula 2 can be rewritten as the following formula 3:
接著，還可於該第一畫幅、該第二畫幅及該第三畫幅各自的該第一區域周圍定義一第二區域，其中該第二區域大於該第一區域。換句話說，即將該第二區域定義為
參考第4圖所示，於該第一畫幅間的該第二區域內，搜索與該第二畫幅的該第一區域中影像相似的區域，以計算一第一相似度。換句 話說，以該第二畫幅的該第一區域中影像為基礎，對該第一畫幅間的該R區域內搜索與該第二畫幅的該第一區域中影像相似的區域，並判斷其相似程度，例如在第4圖中，S1、S2、S3、S4、S5、S6可分別用於表示一中間畫幅與其前三畫幅、前二畫幅、前一畫幅、後一畫幅、後二畫幅、後三畫幅的相似度。 Referring to FIG. 4 , in the second area between the first frames, an area similar to the image in the first area of the second frame is searched to calculate a first similarity. In other words, based on the image in the first area of the second frame, the R area between the first frames is searched for an area similar to the image in the first area of the second frame, and the The degree of similarity, for example, in Figure 4, S 1 , S 2 , S 3 , S 4 , S 5 , S 6 can be used to represent an intermediate frame and its first three frames, the first two frames, the previous frame, and the next frame, respectively. The similarity of the frame, the last two frames, and the last three frames.
接著，參考第4圖所示，還可將該第一相似度換為一差異度，進而求得該滾盪機率，從而判斷該第一畫幅的該第二區域內的滾盪程度，可用於作為該目標影像在該感興趣區的狀態指標，在此例中，該狀態指標可用於指示液面滾盪狀態，但不以此為限。 Next, referring to Fig. 4, the first similarity can also be replaced by a difference, and then the rolling probability can be obtained, so as to determine the rolling degree in the second area of the first frame, which can be used for As the state indicator of the target image in the region of interest, in this example, the state indicator can be used to indicate the state of liquid surface tumbling, but not limited thereto.
可選地，還可於該第三畫幅的該第二區域內，搜索與該第二畫幅的該第一區域中影像相似的區域，以計算一第二相似度；及將該第二相似度轉換為另一差異度後，進而求得另一滾盪機率，從而判斷第三畫幅的該第二區域內的滾盪程度。依此類推，還可將n(例如n可為大於2的正整數)張畫幅以如上所述的方式比對該第二區域的相似度。 Optionally, in the second area of the third frame, search for an area similar to the image in the first area of the second frame to calculate a second similarity; and the second similarity After being converted into another degree of difference, another rolling probability is obtained, thereby judging the degree of rolling in the second area of the third frame. By analogy, n (for example, n may be a positive integer greater than 2) frames can be compared for the similarity of the second region in the manner described above.
具體地，於前後畫幅的第二區域內搜索與該第一畫幅的該第一區域中影像相似的區域，上述式3可以改寫為下式4表示： Specifically, searching for an area similar to the image in the first area of the first frame in the second area of the front and rear frames, the above formula 3 can be rewritten as the following formula 4 to express:
上述的計算方式可以如第5圖所示，其中一個畫幅(例如該第一畫幅)的P區間(例如該第一區域)，在另一個畫幅(例如該第二畫幅)的R區間(例如該第二區域)中搜尋其最大相似度的區域。上述計算方式可以針對前後連續的多個時間區段的圖像進行判斷。 The above calculation method can be as shown in Figure 5, in which the P interval (such as the first area) of one frame (such as the first frame) is in the R interval (such as the second frame) of another frame (such as the second frame). The second area) is searched for the area with the greatest similarity. The above calculation method can be used to judge the images of multiple consecutive time segments before and after.
可選地，在一實施例中，還可於該第一畫幅、該第二畫幅及該第三畫幅各自的一第三區域周圍定義一第四區域，其中該第四區域大於該第三區域，例如該第三區域與該第一區域不重疊，及該第四區域與該第 二區域不重疊；還可於該第一畫幅間的該第四區域內，搜索與該第二畫幅的該第三區域中影像相似的區域，以計算一第三相似度；還可於該第三畫幅間的該第四區域內，搜索與該第二畫幅的該第三區域中影像相似的區域，以計算一第四相似度；及可將該第三相似度及該第四相似度轉換為另一差異度後，進而求得另一滾盪機率，從而判斷該第四區域內的另一滾盪程度，進而對該模液面的整個液面進行滾盪監測。 Optionally, in one embodiment, a fourth area may also be defined around a third area of each of the first frame, the second frame and the third frame, wherein the fourth area is larger than the third area , for example, the third area does not overlap with the first area, and the fourth area and the The two areas do not overlap; the fourth area between the first frames can also be searched for an area similar to the image in the third area of the second frame to calculate a third similarity; In the fourth area between the three frames, searching for an area similar to the image in the third area of the second frame to calculate a fourth similarity; and converting the third similarity and the fourth similarity After being another difference degree, another tumbling probability is obtained, so as to judge another tumbling degree in the fourth area, and then monitor the tumbling of the entire liquid surface of the mold.
最後，可將該第一相似度換為一差異度後，進而求得一滾盪機率(可作為一差異機率)，從而判斷該第二區域內的一滾盪程度(可作為一差異程度)，作為該目標影像在該感興趣區的狀態指標，可用於指示該感興趣區隨時間產生的差異狀態，進而可供人員或機器進行相關分析或評估作業的依據。 Finally, after replacing the first similarity with a difference, a rolling probability (which can be used as a difference probability) can be obtained, so as to determine a rolling degree in the second area (which can be used as a difference degree) , as the state indicator of the target image in the region of interest, which can be used to indicate the difference state of the region of interest over time, which can then be used as a basis for personnel or machines to perform related analysis or evaluation operations.
可選地，還可將該滾盪程度高的區域進行標示。請參考第6圖所示，經過連續的比對及連續計算處理後，可以得到連續時間滾盪數據，例如橫軸代表連續畫幅順序，縱軸代表滾盪程度。 Optionally, the area with a high degree of rolling can also be marked. Please refer to Figure 6. After continuous comparison and continuous calculation, continuous time rolling data can be obtained. For example, the horizontal axis represents the sequence of continuous frames, and the vertical axis represents the degree of rolling.
將第6圖得到的連續時間滾盪數據經過數值分析處理(例如window averaging)後，可以將滾盪機率轉換成如第7圖所示的曲線圖，例如橫軸代表時間(單位為秒)，縱軸代表滾盪程度，其中左邊方框(例如表示第一畫幅的感興趣區)代表該時段的滾盪機率較低，右邊方框(例如表示第三畫幅的感興趣區)代表該時段的滾盪機率較高。換句話說，該第三畫幅的該第二區域有滾盪機率增高的傾向。 After the continuous time roll data obtained in Figure 6 is subjected to numerical analysis processing (such as window averaging), the roll probability can be converted into a graph as shown in Figure 7, for example, the horizontal axis represents time (in seconds), The vertical axis represents the degree of roll, in which the left box (for example, the ROI of the first frame) represents the lower probability of rolling in this period, and the right box (for example, the ROI of the third frame) represents the period of time. The chance of rolling is higher. In other words, the second region of the third frame tends to have an increased rolling probability.
應被注意的是，由於液面滾盪為一連續現象，而且在滾盪時液面的擾動多會維持一段時間長度，因而可以利用該狀態指標的大數據(例如數值大小以及較高數值發生頻率等)，作為預警是否有滾盪(即連續影像中的液面產生差異)的情況發生。 It should be noted that since the tumbling of the liquid level is a continuous phenomenon, and the disturbance of the liquid level during tumbling will last for a period of time, it is possible to use the big data of this state indicator (such as the magnitude of the value and the occurrence of higher values). frequency, etc.), as an early warning if tumbling (i.e. differences in liquid level in successive images) occurs.
由於上述計算方式牽涉估測區域(如感興趣區)的平面運算，所以當估測區間過大會造成局部變化影響過小，使得數值差異不夠突顯滾盪狀態；反而言之，當估測區間過小，微小變動就會造成極大的數值變化，而使得系統過度敏感。所以比對區間的大小是會影響敏感度。因此，本發明的實施例還可以根據需求調整估測區域的大小。舉例而言，當需要 作大範圍的運算偵測，需要將所運算的區間佈滿相同大小的運算區塊進行獨立運算，得到各局部區域的差異數值。避免以過大的運算區域進行運算，減低對差異(如滾盪)變化的敏感度。 Since the above calculation method involves the plane operation of the estimation area (such as the region of interest), when the estimation interval is too large, the impact of local changes is too small, so that the numerical difference is not enough to highlight the rolling state; on the contrary, when the estimation interval is too small, Small changes can cause large numerical changes, making the system overly sensitive. Therefore, the size of the comparison interval will affect the sensitivity. Therefore, the embodiment of the present invention can also adjust the size of the estimation area according to requirements. For example, when it is necessary to For large-scale operation detection, it is necessary to perform independent operation on the operation area covered with the same size operation block to obtain the difference value of each local area. Avoid operating with an excessively large operating area to reduce sensitivity to variance (such as rollover) changes.
另一方面，本發明還提供一種電腦程式產品，當電腦載入該電腦程式並執行後，該電腦能夠執行如上所述之液面狀態估測方法。例如：該電腦程式產品可包含數個程式指令，該程式指令可利用現有的程式語言(如C或Python等)實現，以便用於執行如上所述之液面狀態估測方法，惟不以此為限。 On the other hand, the present invention also provides a computer program product. After the computer program is loaded and executed, the computer can execute the liquid level state estimation method as described above. For example: the computer program product may contain several program instructions, which may be implemented using existing programming languages (such as C or Python, etc.), so as to be used to implement the liquid level state estimation method as described above, but not limited.
另一方面，本發明還提供一種電腦可讀取紀錄媒體，例如：光碟、隨身碟或硬碟等，該電腦可讀取紀錄媒體內儲程式(如上述電腦程式)，當電腦載入該程式並執行後，該電腦能夠完成如上所述之液面狀態估測方法。 On the other hand, the present invention also provides a computer-readable recording medium, such as an optical disc, a flash drive or a hard disk, etc., the computer can read a program stored in the recording medium (such as the above-mentioned computer program), and when the computer loads the program After execution, the computer can complete the liquid level state estimation method as described above.
承上所述，本發明上述實施例的液面狀態估測方法、電腦程式產品及電腦可讀取紀錄媒體，基於熱影像中的感興趣區定位在易產生捲渣缺陷的區域附近，對相鄰影像的感興趣區位置比對差異，以利有效降低捲渣缺陷。 Based on the above, the liquid level state estimation method, the computer program product and the computer-readable recording medium according to the above-mentioned embodiments of the present invention are based on the positioning of the region of interest in the thermal image near the area prone to slag defects. The differences in the position of the region of interest of the adjacent images are compared to effectively reduce the slag defects.
舉例而言，熱影像的量測區域需要針對邊蓋區域作些微調整，這是因為各個鋼料胚型尺寸不同，需要在不同的產品製程中調整，然而當邊蓋調整時，該量測區域也需要整個移動，其中一個原因是因為原本的量測區域可能會出現在調整後的量測區域外，因而造成量測失效；另一原因是由於量測的感興趣區常會出現在模邊，所以當邊蓋出現調整，該量測區域也需要調整。 For example, the measurement area of the thermal image needs to be adjusted slightly for the side cover area. This is because the size of each steel blank is different and needs to be adjusted in different product processes. However, when the side cover is adjusted, the measurement area is The entire movement is also required. One of the reasons is that the original measurement area may appear outside the adjusted measurement area, resulting in measurement failure; Therefore, when the side cover is adjusted, the measurement area also needs to be adjusted.
因此，本發明上述實施例的液面狀態估測方法、電腦程式產品及電腦可讀取紀錄媒體，通過將該液面區域內的感興趣區定位在該定位基線附近；將該感興趣區映對到該目標影像的至少一相鄰影像的區域設定為至少一比對熱區；及估測該比對熱區與該感興趣區之間的差異程度，依據該差異程度產生該目標影像的狀態指標，可用於煉鋼製程中追蹤邊蓋位置調整該量測區域，利用邊蓋位置定位該量測區域離邊蓋之距離，可以有效偵測較易發生問題的邊蓋週邊情況，以利有效降低煉鋼缺陷。 Therefore, the liquid level state estimation method, the computer program product and the computer-readable recording medium of the above-mentioned embodiments of the present invention locate the region of interest in the liquid level region near the positioning baseline; map the region of interest Setting an area of at least one adjacent image to the target image as at least one comparison hotspot; and estimating the degree of difference between the comparison hotspot and the region of interest, and generating a comparison of the target image according to the degree of difference The status indicator can be used to track the position of the side cover and adjust the measurement area in the steelmaking process. Using the position of the side cover to locate the distance between the measurement area and the side cover can effectively detect the surrounding conditions of the side cover that are prone to problems, so as to facilitate Effectively reduce steelmaking defects.
雖然本發明已以較佳實施例揭露，然其並非用以限制本發明，任何熟習此項技藝之人士，在不脫離本發明之精神和範圍內，當可作各種更動與修飾，因此本發明之保護範圍當視後附之申請專利範圍所界定者為準。 Although the present invention has been disclosed with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of protection shall be determined by the scope of the appended patent application.
P1:取像步驟 P1: Image acquisition steps
P2:測邊步驟 P2: Steps for edge measurement
P3:定位步驟 P3: Positioning step
P4:映對步驟 P4: Mapping step
P5:估測步驟 P5: Estimation step
P6:示警步驟 P6: Warning steps
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW109119212A TWI721901B (en) | 2020-06-08 | 2020-06-08 | Method for estimating state of liquid level, computer program product, and computer readable recording medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW109119212A TWI721901B (en) | 2020-06-08 | 2020-06-08 | Method for estimating state of liquid level, computer program product, and computer readable recording medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TWI721901B TWI721901B (en) | 2021-03-11 |
| TW202147251A true TW202147251A (en) | 2021-12-16 |
Family
ID=76036070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW109119212A TWI721901B (en) | 2020-06-08 | 2020-06-08 | Method for estimating state of liquid level, computer program product, and computer readable recording medium |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI721901B (en) |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108318101A (en) * | 2017-12-26 | 2018-07-24 | 北京市水利自动化研究所 | Water gauge water level video intelligent monitoring method based on deep learning algorithm and system |
| CN109443480B (en) * | 2018-11-02 | 2020-05-26 | 南京邮电大学 | Water level scale positioning and water level measuring method based on image processing |
-
2020
- 2020-06-08 TW TW109119212A patent/TWI721901B/en active
Also Published As
| Publication number | Publication date |
|---|---|
| TWI721901B (en) | 2021-03-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hoang et al. | Metaheuristic optimized edge detection for recognition of concrete wall cracks: a comparative study on the performances of roberts, prewitt, canny, and sobel algorithms | |
| US10704900B2 (en) | Detection device and detection method | |
| Zou et al. | Automatic weld defect detection method based on Kalman filtering for real-time radiographic inspection of spiral pipe | |
| TWI600897B (en) | Computer-implemented method, non-transitory computer-readable medium, and system for detecting defects on a wafer | |
| JP4631096B2 (en) | Vehicle periphery monitoring device | |
| WO2016107474A1 (en) | Vehicle checking method and system | |
| KR102021945B1 (en) | Determination of localised quality measurements from a volumetric image record | |
| US20120027288A1 (en) | Methods, Systems and Apparatus for Defect Detection | |
| CN115862259B (en) | Fire alarm early warning system based on temperature monitoring | |
| US10593044B2 (en) | Information processing apparatus, information processing method, and storage medium | |
| WO2021000948A1 (en) | Counterweight weight detection method and system, and acquisition method and system, and crane | |
| US10643338B2 (en) | Object detection device and object detection method | |
| CN103077534B (en) | Spatiotemporal object moving target detecting method | |
| JP6278880B2 (en) | Water level measuring device | |
| US9972095B2 (en) | Image measuring apparatus and non-temporary recording medium on which control program of same apparatus is recorded | |
| CN107808165B (en) | Infrared image matching method based on SUSAN corner detection | |
| JP6221283B2 (en) | Image processing apparatus, image processing method, and image processing program | |
| TWI721901B (en) | Method for estimating state of liquid level, computer program product, and computer readable recording medium | |
| CN113763376B (en) | Conveyor belt offset detection method, conveyor belt offset detection device, computer equipment and storage medium | |
| US20140297224A1 (en) | Data processing method of tire shape inspecting device | |
| US11796481B2 (en) | Inspection device and inspection method | |
| TWI757590B (en) | Luminance uniform detection system and luminance uniform detection method | |
| JP7405362B2 (en) | Concrete structure diagnosis system, concrete structure diagnosis method and program | |
| JP2021064215A (en) | Surface property inspection device and surface property inspection method | |
| CN116608816B (en) | Calibration method and device for calibrating device of small-angle measuring instrument |