Cinder ladle water flow detection method and system

Abstract

The invention belongs to the technical field of steel smelting, and discloses a slag ladle water flow detection method and system. The method comprises the following steps: respectively acquiring reference information of an infrared image detection area and reference information of a visible light image detection area based on an infrared image and a visible light image of a cinder ladle water flow area in a water-flow-free state; obtaining an infrared image and a visible light image of the cinder ladle water flow area in a to-be-detected state, and performing denoising and correction processing respectively to obtain a to-be-detected infrared image and a to-be-detected visible light image; the to-be-detected infrared image is matched with the reference information of the infrared image detection area, and the to-be-detected visible light image is matched with the reference information of the visible light image detection area; for the reference information of the to-be-detected infrared image and the infrared image detection area and the reference information of the to-be-detected visible light image and the visible light image detection area, respectively extracting water flow features by adopting a background difference method; and judging the cinder ladle water flow condition according to the water flow characteristics. The device is high in water flow detection accuracy and good in stability.

Description

Technical Field

[0001] The invention belongs to the technical field of steel smelting, and in particular relates to a slag ladle water flow detection method and system. Background Art

[0002] In the industrial production process, especially in the steel smelting industry, the slag ladle needs to be watered to cool it down. The water flow detection of the slag ladle is a key link in the slag ladle treatment process. Accurate detection of the water flow of the slag ladle is of great significance to ensure production safety and improve production efficiency.

[0003] The conventional method for detecting water flow in slag ladle is to use a temperature sensor based on a thermal resistor or a thermocouple to detect the temperature change on the surface of the slag ladle, thereby indirectly judging the water flow situation. However, this method has limited collection points and cannot fully reflect the water flow situation on the surface of the slag ladle, which is prone to misjudgment and missed judgment; and due to the complex environment of the slag ladle, the single method of indirectly detecting water flow through the slag ladle temperature is easily affected by the ambient temperature, resulting in inaccurate detection results.

[0004] The conventional water flow detection method is to set a liquid level sensor at the water source to detect the water flow. However, this method is not suitable for slag bag water flow detection and cannot accurately detect the situation where water flows to the slag bag.

[0005] Therefore, the above methods cannot meet the requirements for slag ladle water flow detection accuracy and detection stability in existing production. Summary of the invention

[0006] In order to solve the above problems, one object of the present invention is to provide a slag ladle water flow detection method, which has the characteristics of high detection accuracy and good stability.

[0007] The specific technical solutions are as follows: A method for detecting water flow in a slag ladle comprises the following steps: S1. Based on the infrared image and visible light image of the slag ladle water flow area in the non-water flow state, obtain the reference information of the infrared image detection area and the reference information of the visible light image detection area respectively; S2. Obtain an infrared image and a visible light image of the slag ladle water flow area in the state to be tested, and perform denoising and correction processing respectively to obtain an infrared image to be tested and a visible light image to be tested; the infrared image to be tested matches the reference information of the infrared image detection area, and the visible light image to be tested matches the reference information of the visible light image detection area; S3. Based on the reference information of the infrared image to be tested and the infrared image detection area, and the reference information of the visible light image to be tested and the visible light image detection area, respectively, the water flow characteristics are extracted by using the background difference method; and the water flow condition of the slag ladle is determined according to the water flow characteristics.

[0008] Furthermore, in the step S1, the detection area is located and marked on the infrared image and the visible light image respectively, and the obtained position information and the marked area image are used as reference information.

[0009] Furthermore, in step S2, the correction processing specifically includes: respectively marking the infrared image and the visible light image of the slag ladle water flow area in the measured state to obtain the corresponding infrared image marking area and the visible light image marking area; respectively aligning the marked area with the corresponding reference information; intercepting the aligned marked area for affine transformation, and using the perspective transformation method to realize image correction.

[0010] Furthermore, the step S3 specifically includes: S31. The infrared image to be tested and the visible light image to be tested are used as the images to be tested, and steps S311 to S314 are respectively executed to obtain the corresponding alarm signal; S311. Performing differential processing on the image to be tested and the image of the marked area in the corresponding reference information to obtain a differential image; S312. De-noising the differential image; S313. Binarizing the denoised differential image to obtain a binary image and then denoising it; S314. Using morphological contour detection method, count the number of contours and corresponding contour information contained in the binary image, detect the position range of each contour, set the range limit value, and exclude the interference of non-water flow area; calculate the row and column cumulative values ​​of the processed binary image, and determine whether the cumulative value is within the preset range. If it exceeds the preset range, output an alarm signal; S32. Determine the water flow condition in the image to be tested based on the alarm signal corresponding to the infrared image to be tested and the alarm signal corresponding to the visible light image to be tested.

[0011] Furthermore, the step S312 specifically includes: According to the morphological characteristics of the slag bag water flow, the morphological erosion and dilation operation is used to perform morphological filtering operations to remove the interference information and small noise points contained in the differential image. Set the area threshold to filter out small connected domains in the difference image.

[0012] Furthermore, the method also includes: obtaining the temperature distribution of the slag bag through infrared data to determine whether the temperature of the slag bag is abnormal; combining the slag bag temperature determination result with the water flow determination results of the infrared image and the visible light image to jointly determine whether the water flow corresponding to the slag bag is abnormal.

[0013] Furthermore, the method of obtaining the temperature distribution of the slag bag through infrared data and judging whether the temperature of the slag bag is abnormal specifically includes: collecting actual temperature data of the slag bags in the same group during the cooling process, wherein the slag bags in the same group are multiple slag bags that start watering and cooling at the same time; The temperature change trend of the slag bags in the same group is obtained by data fitting; the mean value of the temperature change trend of the slag bags in the same group is calculated according to the temperature change trend; the absolute deviation is calculated according to the temperature change trend corresponding to the slag bags in the same group and the mean value of the temperature change trend, and the absolute deviation is compared with the preset value to determine whether the current slag bag temperature is abnormal. Furthermore, when there are two or more abnormal judgment results in the judgment results of the slag bag temperature, infrared image, and visible light image, it is determined that the slag bag water flow is abnormal.

[0014] Another object of the present invention is to provide a slag ladle water flow detection system, which applies the above-mentioned slag ladle water flow detection method, and the system comprises: Visible light module, used to obtain visible light images of the water flow area of ​​the slag ladle; Infrared movement, used to obtain infrared images of the slag ladle water flow area; A network module, used for transmitting the visible light image and the infrared image to a processing module; The processing module is used to extract water flow characteristics based on the visible light image and infrared image of the water flow area of ​​the slag bag, and judge the water flow situation based on the water flow characteristics.

[0015] Furthermore, the system also includes an infrared thermometer for obtaining infrared data of the slag ladle surface; the network module is also used to convert the infrared data of the slag ladle surface into temperature information and transmit it to the processing module; the processing module extracts water flow characteristics based on the visible light image and infrared image of the water flow area of ​​the slag ladle, and determines the water flow conditions based on the water flow characteristics and temperature information.

[0016] Compared with the prior art, one or more of the above technical solutions can achieve at least one of the following beneficial effects: The present invention adopts a method of combining multi-dimensional information to improve the accuracy and stability of slag ladle water flow detection and reduce misjudgment and missed judgment. Use feature information and infrared temperature data to establish a detection model to reduce the impact of environmental interference on the detection results and improve the reliability of detection; It can realize remote real-time detection and judgment of slag bag water flow, improving production efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a schematic flow chart of the slag ladle water flow detection method of Example 1.

[0019] Figure 2 Schematic diagram of detection area marking in Example 1; wherein, Figure (a) is a schematic diagram of detection area marking in visible light image, and Figure (b) is a schematic diagram of detection area marking in infrared image.

[0020] Figure 3 The figures are images of different steps in the image preprocessing process of Example 1; wherein, Figure (c) is the image of the marked area in the reference information, Figure (d) is the image to be tested after preprocessing, Figure (e) is the binary image after difference, and Figure (f) is the image after morphological processing.

[0021] Figure 4 This is a principle block diagram of the slag ladle water flow detection system of Example 3. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Example 1 like Figure 1 As shown, this embodiment provides a slag ladle water flow detection method, comprising the following steps: S1. Water flow detection area calibration: Based on the infrared image and visible light image of the slag ladle water flow area in the non-water flow state, obtain the reference information of the infrared image detection area and the reference information of the visible light image detection area respectively.

[0024] Specifically, Figure 2 As shown, the detection area is located and marked in the infrared image and the visible light image respectively, and the obtained position information and the marked area image are saved as reference information.

[0025] S2. Image preprocessing: Obtain infrared images and visible light images of the slag ladle water flow area in the test state, and perform denoising and correction processing respectively to obtain the infrared image to be tested and the visible light image to be tested; the infrared image to be tested is matched with the baseline information of the infrared image detection area, and the visible light image to be tested is matched with the baseline information of the visible light image detection area.

[0026] Specifically, the infrared image and visible light image of the slag ladle water flow area in the state to be tested are respectively annotated to obtain the corresponding infrared image annotated area and visible light image annotated area; the infrared image annotated area and the reference information of the infrared image detection area are registered, and the visible light image annotated area and the reference information of the visible light detection area are registered. Since the image scene information of the image to be tested and the annotated area of ​​the reference information are basically the same, and there are only differences in translation, rotation, etc. between the two, the registration in this algorithm adopts the region-based image registration method.

[0027] The image of the registration area is intercepted for radial transformation, and the image correction is realized by using the perspective transformation method to ensure that the size and shape of the regional image are consistent with the reference image. The essence of perspective transformation is to project the image to a new viewing plane. Its general transformation formula is: (x, y, z) is the original coordinate, (X, Y, Z) is the transformed coordinate, is the pixel coordinate of the image after transformation, where ; The perspective transformation matrix diagram is as follows: Represents a linear transformation of an image.

[0028] Used to generate image perspective transformation.

[0029] Indicates image translation.

[0030] S3. Water flow detection: Figure 3 As shown, the background difference method is used to extract water flow characteristics for the infrared image to be tested and the reference information of the infrared image detection area, and the reference information of the visible light image to be tested and the visible light image detection area; and the water flow situation of the slag ladle is judged according to the water flow characteristics.

[0031] S31. The infrared image to be tested and the visible light image to be tested are used as the images to be tested, and steps S311 to S314 are respectively executed to obtain the corresponding alarm signal; S311. Perform differential processing to obtain a differential image: Since the water flow area is basically fixed, the image to be tested and the reference information image can be first registered and corrected, and the processed image to be tested and the marked area image in the corresponding reference information can be differentially processed; the currently acquired image is subtracted from the previously stored background image, and the area that deviates from the background by more than a certain threshold is calculated as the water flow area.

[0032] S312. Perform denoising on the differential image.

[0033] According to the morphological characteristics of the slag bag water flow, the morphological erosion and dilation operation is used to perform morphological filtering operations to remove small noise points and small interferences contained in the differential image. The morphological structure operator used here is a 4*3 structure operator.

[0034] Set the area threshold to filter out small connected domains and eliminate the interference of small area noise. Through denoising, reduce the influence of external light, equipment jitter and other factors in the actual detection process.

[0035] S313. Binarize the denoised differential image, and perform denoising again after obtaining a binary image.

[0036] S314. Use morphological contour detection to count the number of contours and corresponding contour information contained in the binary image, detect the position range of each contour, set the range limit value, and exclude the interference of non-water flow areas; calculate the row and column cumulative values ​​of the processed binary image, and determine whether the column cumulative values ​​of the continuous n1 columns of binary images and the row cumulative values ​​of the continuous n2 rows of binary images in the image are both within the range of [n3, n4]. If not, output an alarm message. Among them, n1, n2, n3, and n4 are empirical thresholds.

[0037] S32. Through steps S311 to S314, water flow information is identified on the infrared image and the visible light image respectively to obtain the infrared image alarm result. And the light image alarm result According to the alarm signal corresponding to the infrared image to be tested and the alarm signal corresponding to the visible light image to be tested, the water flow situation in the image to be tested is jointly determined to determine whether the water flow is interrupted.

[0038] This embodiment adopts the method of multi-dimensional information fusion of infrared light and visible light, which overcomes the defect that single visible light detection is easily affected by ambient light and shadow, resulting in unstable detection accuracy; it overcomes the defect that single infrared light detection is easily affected by ambient temperature and slag bag surface material, resulting in inaccurate detection results. This method is particularly suitable for complex slag bag production environments, with strong anti-interference ability, high detection accuracy and good stability.

[0039] Example 2 The difference between this embodiment and embodiment 1 is that, on the basis of embodiment 1, a method for temperature measurement and water flow prediction is added: the temperature distribution of the slag bag is obtained through infrared images to determine whether the temperature of the slag bag is abnormal; the slag bag temperature judgment result is combined with the judgment results of infrared images and visible light images to jointly determine whether the water flow of the slag bag is abnormal. Since the cooling rate of the slag bag is quite different when it is watered and not watered (natural cooling), and if the water flow is interrupted during the cooling process of the slag bag, the temperature of the slag bag will have a short temperature recovery process. Based on this feature, the water flow is detected by the slag bag temperature side.

[0040] The actual temperature distribution of the slag bag during the cooling process is collected using an infrared thermal imager. In the specific implementation, the actual temperature data of the slag bags in the same group during the cooling process are collected, wherein the slag bags in the same group are multiple slag bags that start watering and cooling at the same time.

[0041] The trend rate TempR of the slag bag temperature change is obtained by data fitting. In actual production, since the slag bags in the same group are generally watered and cooled at the same time, the trend rate TempR of the temperature change will remain consistent. Taking a group of 8 slag bags as an example, the temperature change trend rate of a group of 8 slag bags is processed as follows: Get trend array : in, to Respectively represent the temperature change trends of 8 slag bags.

[0042] Calculate the mean of the temperature change trend of the same group of slag bags according to the temperature change trend, that is, the array mean : According to the temperature change trend and the mean value of the temperature change trend of the slag bags in the same group, the absolute deviation of the temperature change trend of each slag bag is calculated, that is, the absolute deviation of the array. : Compare the absolute deviation of each slag bag with the preset value to determine whether the temperature of each slag bag is abnormal: in, It is a preset value. Here, the preset value is an empirical value. 0 means that the water flow is cut off, and 1 means that the water flow is normal.

[0043] When there are two or more abnormal judgment results among the judgment results of the temperature, infrared image and visible light image of a slag bag, it is judged that the water flow of the slag bag is abnormal.

[0044] in It is the alarm sign of infrared image; It is an alarm sign of visible light image; It is the alarm sign of the temperature measurement result.

[0045] The slag ladle water flow detection method of this embodiment uses infrared temperature data and characteristic information to establish a water flow detection model, which overcomes the defect that the detection result is inaccurate due to the complex slag ladle environment and the single temperature measurement method is easily affected by the air temperature and the surface material of the slag ladle. The method of this embodiment further improves the accuracy of the detection information.

[0046] Example 3 like Figure 4 As shown, the embodiment provides a slag ladle water flow detection system, which applies the slag ladle water flow detection method of embodiment 2, and the system includes: Visible light module, used to obtain visible light images of the water flow area of ​​the slag ladle; Infrared movement, used to obtain infrared images of the slag ladle water flow area; Infrared thermometer, used to obtain infrared data of the slag ladle surface; A network module, used for transmitting the visible light image and the infrared image to the processing module; The processing module (not shown in the figure) is used to extract water flow characteristics based on visible light images and infrared images, and combine the water flow characteristics and infrared data to determine the actual water flow conditions and issue warnings.

[0047] As a specific implementation, the visible light module and the infrared movement can adopt a dual-light gimbal that integrates the two.

[0048] As an implementation mode, the infrared thermometer can transmit infrared data to the processing module. As another implementation mode, the infrared data of the infrared thermometer can be converted into temperature information through the network module and then transmitted to the processing module.

[0049] Among them, the parameters of the visible light module, infrared movement, and network module are shown in Table 1, Table 2, and Table 3 respectively.

[0050] Table 1 Visible light module parameters Table 2 Infrared movement parameters Table 3 Network module parameters In the specific implementation process, the processing module can use the server in the control pan-tilt. The dual-light pan-tilt patrols the slag ladle water flow area at a fixed position; the infrared image, visible light image and infrared temperature measurement data of each patrol position are transmitted to the server of the control pan-tilt, and the slag ladle water flow detection model constructed based on the detection method process of Example 2 is run on the server.

[0051] This embodiment can realize remote automatic detection of the water flow of the slag bag at the work site.

[0052] Obviously, the above embodiments are merely examples for clearly explaining the technical solutions of the present invention, and are not intended to limit the implementation methods of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the protection of the claims of the present invention.

Claims

1. A method for detecting water flow in a slag ladle, characterized in that: The following steps are involved: S1. Based on the infrared image and visible light image of the slag ladle water flow area in the non-water flow state, obtain the reference information of the infrared image detection area and the reference information of the visible light image detection area respectively; S2. Obtain an infrared image and a visible light image of the slag ladle water flow area in the state to be tested, and perform denoising and correction processing respectively to obtain an infrared image to be tested and a visible light image to be tested; the infrared image to be tested matches the reference information of the infrared image detection area, and the visible light image to be tested matches the reference information of the visible light image detection area; S3. Extracting water flow features by using background difference method for the reference information of the infrared image to be tested and the infrared image detection area, and the reference information of the visible light image to be tested and the visible light image detection area; Determine the water flow condition of the slag ladle based on the water flow characteristics.

2. The slag ladle water flow detection method according to claim 1, characterized in that: In the step S1, the detection area is located and marked on the infrared image and the visible light image respectively, and the obtained position information and the marked area image are used as reference information.

3. The slag ladle water flow detection method according to claim 1, characterized in that: In step S2, the correction process specifically includes: marking the infrared image and visible light image of the slag ladle water flow area in the test state to obtain corresponding infrared image marking areas and visible light image marking areas; respectively aligning the marked areas with corresponding reference information; intercepting the aligned marked areas for affine transformation, and using perspective transformation to achieve image correction.

4. The slag ladle water flow detection method according to claim 2, characterized in that: The step S3 specifically includes: S31. The infrared image to be tested and the visible light image to be tested are used as the images to be tested, and steps S311 to S314 are respectively executed to obtain the corresponding alarm signal; S311. Performing differential processing on the image to be tested and the image of the marked area in the corresponding reference information to obtain a differential image; S312. De-noising the differential image; S313. Binarizing the denoised differential image to obtain a binary image and then denoising it; S314. Using morphological contour detection method, count the number of contours and corresponding contour information contained in the binary image, detect the position range of each contour, set the range limit value, and exclude the interference of non-water flow area; calculate the row and column cumulative values ​​of the processed binary image, and determine whether the cumulative value is within the preset range. If it exceeds the preset range, output an alarm signal; S32. Determine the water flow condition in the image to be tested based on the alarm signal corresponding to the infrared image to be tested and the alarm signal corresponding to the visible light image to be tested.

5. The slag ladle water flow detection method according to claim 4, characterized in that: The step S312 specifically includes: According to the morphological characteristics of the slag bag water flow, the morphological erosion and dilation operation is used to perform morphological filtering operations to remove the interference information and small noise points contained in the differential image. Set the area threshold to filter out small connected domains in the difference image.

6. The slag ladle water flow detection method according to any one of claims 1 to 5, characterized in that: The method further includes: obtaining the temperature distribution of the slag bag through infrared data to determine whether the temperature of the slag bag is abnormal; and determining whether the water flow corresponding to the slag bag is abnormal by combining the slag bag temperature determination result with the water flow determination result of the infrared image and the visible light image.

7. The slag ladle water flow detection method according to claim 6, characterized in that: The method for obtaining the temperature distribution of slag bags through infrared data and judging whether the temperature of the slag bags is abnormal specifically includes: collecting actual temperature data of slag bags in the same group during the cooling process, wherein the slag bags in the same group are multiple slag bags that start watering and cooling at the same time; obtaining the temperature change trend of the slag bags in the same group by data fitting; calculating the mean value of the temperature change trend of the slag bags in the same group according to the temperature change trend; calculating the absolute deviation of the temperature change trend of each slag bag according to the temperature change trend corresponding to the slag bags in the same group and the mean value of the temperature change trend, comparing the absolute deviation with a preset value, and judging whether the current slag bag temperature is abnormal.

8. The slag ladle water flow detection method according to claim 6, characterized in that: When there are two or more abnormal judgment results among the judgment results of the slag ladle temperature, infrared image, and visible light image, it is judged that the slag ladle water flow is abnormal.

9. A slag ladle water flow detection system, characterized in that: The slag ladle water flow detection method according to any one of claims 1 to 8 is applied, and the system comprises: Visible light module, used to obtain visible light images of the water flow area of ​​the slag ladle; Infrared movement, used to obtain infrared images of the slag ladle water flow area; A network module, used for transmitting the visible light image and the infrared image to a processing module; The processing module is used to extract water flow characteristics based on the visible light image and infrared image of the water flow area of ​​the slag bag, and judge the water flow situation based on the water flow characteristics.

10. The slag ladle water flow detection system according to claim 9, characterized in that: The system also includes an infrared thermometer for obtaining infrared data of the slag ladle surface; the network module is also used to convert the infrared data of the slag ladle surface into temperature information and transmit it to the processing module; the processing module is used to extract water flow characteristics based on visible light images and infrared images of the water flow area of ​​the slag ladle, and judge the actual water flow conditions based on the water flow characteristics and temperature information.

Images