A method and device for detecting the uniformity of paint spraying in the production of door and window frames

Abstract

The application discloses a detection method and equipment for paint spraying uniformity in a door and window frame production process, and belongs to the technical field of image detection. The method comprises the following steps: image acquisition is performed on the outer wall surface of a door and window frame workpiece ending the paint spraying process, and straight line detection is performed on the obtained outer wall surface image to obtain a frame image; macro uniformity features are identified in the image after the frame image is subjected to grayscale processing; if the door and window frame workpiece does not have the macro uniformity features, a preset heat source excitation is applied to the outer wall surface of the workpiece, a frame sequence thermal image is obtained through thermal image and frame image matching; frame splitting is performed on the frame sequence thermal image, and temperature change curves of each component are generated according to the splitting result; a temperature superposition curve is obtained according to the temperature change curves, so as to judge the paint spraying uniformity of the door and window frame workpiece through the temperature superposition curve. The application realizes the automatic detection of the paint spraying uniformity of the door and window frame workpiece through the above method, and guarantees the accuracy and precision of the detection result.

Description

Technical Field

[0001] This application relates to the field of image detection technology, and in particular to a method and equipment for detecting the uniformity of paint spraying during the production of door and window frames. Background Technology

[0002] Door and window frames, as a commonly used building decoration material, undergo multiple processes during production. The painting process, in particular, has a strict procedure to ensure uniform paint application. Especially for commonly used fire-resistant and insulated doors and windows, an additional layer of fire-resistant and heat-insulating material needs to be applied between the primer and topcoat. These types of door and window frames have even higher requirements for paint uniformity. Uneven paint layers will affect the adhesion of subsequent layers, leading to delamination and peeling during later use. Furthermore, it will affect fire resistance and heat insulation performance. This is because the exterior walls of door and window frames are exposed to wind, rain, and sunlight for extended periods; uneven paint layers will result in different areas of the window frame having varying heat insulation properties, impacting user experience and reducing the lifespan of the door and window frames. Therefore, testing the uniformity of paint application on door and window frames is crucial.

[0003] Current technologies mostly focus on improving painting equipment, aiming to achieve uniform painting by adjusting parameters such as the spraying angle and trajectory. However, in multi-layer painting scenarios, the final painting result lacks inspection. For example, when painting door and window frames on hoisting brackets, paint may fall or drip, affecting not only the aesthetics but also the uniformity. The lack of a post-painting inspection process can easily lead to missed inspections, allowing workpieces with poor paint uniformity to leave the factory.

[0004] Some door and window frame manufacturers conduct manual inspections during or after the drying process of the painted frames. However, the naked eye can only observe macroscopic problems such as cracks and bubbles. Issues like uneven paint thickness leading to uniformity are difficult to detect, especially since the exterior wall color of the door and window frames is uniform before leaving the factory, further increasing the difficulty of manual uniformity inspection and resulting in low accuracy. Furthermore, if these door and window frames with uneven paint are installed and used normally, they are prone to problems such as poor heat insulation, cracking, and peeling during later use. This increases maintenance costs and shortens the lifespan of the frames, leading to resource waste. Summary of the Invention

[0005] This application provides a method and equipment for detecting the uniformity of paint spraying during the production of door and window frames, which is used to solve at least one of the above-mentioned technical problems.

[0006] The technical solution adopted in this application is as follows:

[0007] On one hand, this application provides a method for detecting the uniformity of paint spraying during the production of door and window frames. The method includes: acquiring an image of the exterior wall surface of a door and window frame workpiece after the painting process has ended, and performing line detection on the acquired exterior wall surface image to obtain a frame image corresponding to the door and window frame workpiece based on the line detection result; performing grayscale processing on the frame image to identify the macroscopic uniformity features of the door and window frame workpiece in the obtained grayscale image, wherein the macroscopic uniformity features include at least one of missed paint spraying, over-painting, peeling, and blistering; applying a preset heat source excitation to the exterior wall surface of the door and window frame workpiece that does not possess the macroscopic uniformity features, and acquiring a thermal image of the door and window frame workpiece using a thermal imager, and performing image matching on the thermal image using the frame image to obtain a frame sequence thermal image of the door and window frame workpiece; and using the polygonal shape corresponding to the door and window frame workpiece as a reference, processing the... The frame sequence heat map is split into components, and the average temperature sequence corresponding to each component sequence heat map is calculated. Temperature change curves corresponding to each component of the door / window frame workpiece are plotted based on these average temperature sequences. These temperature change curves characterize the correlation between the average temperature of each component of the door / window frame workpiece and time. The time of the temperature change curves corresponding to each component of the door / window frame workpiece is aligned with the time of the preset heat source excitation signal curve. The temperature change curves are then moved to the layer containing the preset heat source excitation signal curve to obtain a temperature superposition curve. If the temperature superposition curves overlap into a single curve, the paint uniformity of the door / window frame workpiece is determined to be high. If at least one jump curve exists in the temperature superposition curve, the paint uniformity of the component corresponding to the jump curve in the door / window frame workpiece is determined to be low.

[0008] In one possible implementation of this application, identifying the macroscopic uniformity features of the door and window frame workpiece in the obtained grayscale image includes: generating a difference calculation window in the grayscale image based on the size information of each component in the door and window frame workpiece, wherein the length of the difference calculation window is a preset proportion of the length of each component; traversing the grayscale image using the difference calculation window, performing backward difference between the previous window region and the next window region corresponding to the difference calculation window, and replacing the original grayscale value in the previous window region with the value after difference calculation to obtain an enhanced image of the grayscale image; calculating the grayscale histogram corresponding to the enhanced image, and determining the target grayscale value of the enhanced image in the grayscale histogram; adjusting the grayscale of the enhanced image using the target grayscale value to obtain a feature judgment image, so as to identify the macroscopic uniformity features of the door and window frame workpiece in the feature judgment image.

[0009] In one possible implementation of this application, after determining the target grayscale value of the enhanced image in the grayscale histogram, the method further includes: if the target grayscale value is zero, then determining that the macroscopic uniformity feature does not exist in the door and window frame workpiece; if the target grayscale value is not zero, then adjusting the grayscale of the enhanced image using the following formula:

[0010]

[0011] Among them, f ( i 0) f is the grayscale value of the i-th pixel in the enhanced image. ( i ) Let a be the adjusted grayscale value of the i-th pixel, and b be the minimum and maximum values ​​among the target grayscale values, respectively. The feature judgment map obtained after grayscale adjustment is superimposed with the grayscale image and / or the frame image to achieve feature localization of the macroscopic uniformity feature.

[0012] In one possible implementation of this application, the preset heat source excitation adopts a modulated heat source, the output signal curve of which is a sine curve and the output frequency is 0.1Hz.

[0013] In one possible implementation of this application, after obtaining the temperature superposition curve corresponding to the temperature change curve, the method further includes: extracting a set of zero-crossing points from the temperature superposition curve within the time period of the preset heat source excitation, wherein the elements in the set of zero-crossing points are arrays, and each array includes n data points, where n is the number of temperature change curves; calculating the delay time of any array element in the set of zero-crossing points relative to the corresponding zero-crossing point on the sine curve; if the delay time fluctuates within a preset error range, then determining that the temperature superposition curves overlap into one curve; if the delay time has a jump point relative to the preset error range, then determining the curve containing the zero-crossing point corresponding to the jump point as the jump curve.

[0014] In one possible implementation of this application, after determining that the paint uniformity of the component corresponding to the jump curve in the door and window frame workpiece is low, the method further includes: identifying the component corresponding to the jump curve as a non-uniform component, and generating a rectangular test window based on the size information of the non-uniform component, wherein the width of the rectangular test window is equal to the width of the non-uniform component, and the length of the rectangular test window is a preset proportion of the length of the non-uniform component; traversing the frame sequence heatmaps corresponding to the non-uniform component through the rectangular test window; extracting the regions with the same position corresponding to the rectangular test window in the frame sequence heatmap to construct sub-sequence heatmaps, and generating temperature change curves corresponding to the sub-sequence heatmaps respectively; performing time alignment on the temperature change curves corresponding to the sub-sequence heatmaps respectively, so as to determine the separation curve relative to the temperature change curve in the time-aligned curves; and determining the non-uniform paint area in the non-uniform component based on the region position of the rectangular test window corresponding to the separation curve.

[0015] In one possible implementation of this application, after determining the separation curve relative to the temperature change curve, the method further includes: extracting the start separation time node corresponding to the separation curve from the time-aligned curve graph; using a straight line parallel to the time axis as a measurement line, and ensuring that the temperature value corresponding to the measurement line is lower than the temperature value corresponding to the start separation time node; if the measurement line passes through the separation curve first in the time axis direction, it is determined that the paint volume in the uneven paint area is low; if the measurement line passes through the temperature change curve first in the time axis direction and then passes through the separation curve, it is determined that the paint volume in the uneven paint area is high.

[0016] In one possible implementation of this application, traversing the frame sequence heatmaps corresponding to the non-uniform component through the rectangular test window includes: extracting a portion of the sequence heatmaps after the preset heat source excitation stops in the frame sequence heatmaps corresponding to the non-uniform component, traversing the portion of the sequence heatmaps through the rectangular test window, and generating a temperature change curve of the non-uniform component during the cooling phase.

[0017] In one possible implementation of this application, obtaining a frame sequence thermal image of the door and window frame workpiece by image matching the thermal image with the frame image includes: before applying a preset heat source excitation to the door and window frame workpiece, adjusting the image acquisition parameters of the thermal imager and the image acquisition parameters corresponding to the exterior wall image to be consistent, wherein the image acquisition parameters include at least the image acquisition angle, the image acquisition viewing angle, and the image acquisition frame; marking the frame image in the exterior wall image, and superimposing the marked exterior wall image and the thermal image to obtain a frame sequence thermal image corresponding to and matching the frame image in the thermal image.

[0018] On the other hand, this application also provides a device for detecting the uniformity of paint spraying during the production of door and window frames. The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to execute a method for detecting the uniformity of paint spraying during the production of door and window frames as described in any of the above embodiments.

[0019] The method and equipment for detecting the uniformity of paint spraying during the production of door and window frames provided in this application have the following beneficial effects:

[0020] The exterior wall image of the door and window frame workpiece is obtained by image acquisition. Straight lines are extracted from the exterior wall image using line detection technology. The frame image of the door and window frame workpiece is obtained from the straight lines. The frame image is then processed into grayscale. By processing and calculating the obtained grayscale image, the concept of grayscale histogram is used to determine the possible macroscopic uniformity features in the door and window frame workpiece. This method ensures effective identification of macroscopic uniformity features without wasting excessive computing resources, thus saving the cost of identifying macroscopic uniformity features.

[0021] For microscopic uniformity features that are difficult to identify, i.e., differences in uniformity that are not easily detected, this application applies a preset excitation heat source to the outer wall surface of the door and window frame workpiece during identification. This applied excitation heat source allows for the determination of whether microscopic uniformity features exist in the door and window frame workpiece using the obtained thermal image. Furthermore, when the door and window frame workpiece is a fire-resistant and heat-insulating window, it can also perform heat insulation performance testing. During uniformity detection, image matching is first performed using the preceding frame image and thermal image. The frame sequence heat map is segmented from the thermal image. Then, according to the polygonal shape of the door and window frame workpiece, the frame sequence heat map is broken down into sequence heat maps of each component. Using the temperature mean sequence corresponding to the sequence heat map of each component, a temperature change curve corresponding to each component is generated. The temperature change curve and the signal curve corresponding to the preset heat source excitation are time-aligned and moved onto a single image to obtain a temperature superposition curve. Finally, the uniformity of the paint spraying on the door and window frame workpiece can be determined by whether there are any jump curves in the temperature superposition curve. The above process not only enables automated detection of the paint uniformity of door and window frame components, allowing for precise testing of the paint uniformity after painting or before shipment, ensuring that all door and window frame components leaving the factory have high paint uniformity, but also solves the problem of low accuracy caused by traditional manual judgment methods. It also helps to increase the service life of door and window frame components, effectively avoiding the drawbacks such as cracking, paint peeling, and even affecting fireproof and heat insulation performance during use due to low paint uniformity, and also reduces the later maintenance costs of door and window frame components. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0023] Figure 1 A flowchart of a method for detecting the uniformity of paint spraying during the production process of door and window frames, provided in this application;

[0024] Figure 2 Example 1 of the temperature change curve provided for this application;

[0025] Figure 3 This application provides a structural schematic diagram of a device for detecting the uniformity of paint spraying during the production process of door and window frames. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0027] The method in this application will be described in detail below with reference to the accompanying drawings.

[0028] Figure 1 This application provides a flowchart of a method for detecting the uniformity of paint spraying during the production of door and window frames, as shown in the following figure. Figure 1 As shown, the paint uniformity testing method in this application includes at least the following steps:

[0029] Step 101: Acquire images of the exterior wall surface of the door and window frame workpiece after the painting process is completed, and perform line detection on the obtained exterior wall surface image to obtain the frame image.

[0030] Door and window frames undergo multiple processes during production. After the painting process, the door and window frame workpieces are dried before shipment. For fire-resistant and heat-insulating door and window frames, in addition to applying primer and topcoat, a fire-resistant and heat-insulating layer is applied between the two. Each layer requires a drying process before the next layer is applied. Therefore, the paint uniformity testing method in this application can be performed either after all paint layers have dried or after each individual layer has dried. However, to save testing costs and time, it is preferable to perform the test before shipment, i.e., after all paint layers have been painted.

[0031] Specifically, the first step is to acquire images of the exterior wall surface of the door and window frame workpiece. Here, the exterior wall surface refers to the surface of the door and window frame in the outdoor environment. In one example, when acquiring images of the exterior wall surface, the position and parameters of the image acquisition device are adjusted so that the image acquisition device is directly facing the door and window frame workpiece, ensuring that a panoramic image of the door and window frame workpiece can be acquired, that is, all the components of the door and window frame workpiece are captured.

[0032] Furthermore, straight line detection is performed on the acquired exterior wall image. This straight line detection process can be implemented using existing algorithms or methods, which will not be elaborated upon in this embodiment. Since door and window frame components are generally rectangular, there are certain patterns in the straight line detection results. For example, horizontal and vertical lines intersect or there are two sets of opposite horizontal and vertical lines. The detected straight line results are combined to obtain the outline of the door and window frame component. By cropping the exterior wall image according to this outline, the frame image corresponding to the door and window frame component can be obtained, that is, an image containing only the components of the door and window frame component.

[0033] Step 102: Identify the macroscopic uniformity features of the door and window frame workpieces in the image after grayscale processing of the frame image.

[0034] The aforementioned frame image is converted to grayscale to obtain a grayscale image. This grayscale conversion process can be implemented using existing algorithms or methods, which will not be elaborated upon here. The obtained grayscale image is then used to identify macroscopic uniformity features in the middle of the door / window frame. In one example, macroscopic uniformity features refer to easily identifiable low uniformity conditions, including at least one of missed paint application, over-painting, peeling, and blistering.

[0035] In one possible implementation of this application, identifying macroscopic uniformity features in a grayscale image can be achieved through the following process: First, based on the dimensional information of each component in the door / window frame workpiece, such as width or length, a difference calculation window is generated. It is important to note that the length of the difference calculation window needs to be exactly a preset proportion of the length of the door / window frame component, thus ensuring that the difference calculation window can move an integer number of positions when traversing the components. Then, the generated difference calculation window is used to traverse the grayscale image, performing a difference between the previous window position and the next window position. Preferably, this difference is a backward difference, that is, subtracting the pixel value in the next window position from the pixel value in the previous window position, and replacing the original pixel value in the previous window with the result of the subtraction. After the difference is completed, the augmented image corresponding to the grayscale image is obtained. It should be noted that since the difference calculation window is calculated backwards, when the difference calculation window reaches the last position of the component, that is, when one edge of the difference calculation window coincides with one edge of the component and the component is about to be traversed, since there is no corresponding window position area afterward, the difference here can be calculated with the first window position area on the current component, or with the first window position area of ​​the adjacent component of the current component.

[0036] Furthermore, the grayscale histogram corresponding to the enhanced image is calculated, and the target grayscale value of the enhanced image is determined within the histogram. If the paint uniformity of the door and window frame workpiece is high, with no uneven areas or regions, then the target grayscale value obtained here is zero. However, if the target grayscale value obtained here is not zero, it indicates that there are uneven positions or regions in the current door and window frame workpiece. It should be noted that the grayscale image detection used here is for relatively obvious or easily observable unevenness on the door and window frame workpiece, i.e., macroscopic uniformity features. Therefore, the grayscale detection effect is good and the cost is low and easy to implement.

[0037] Furthermore, after determining the existence of macroscopic uniformity features on the door and window frame workpiece based on the target grayscale value, the grayscale of the added image is first adjusted using the following formula:

[0038]

[0039] Among them, f ( i 0) To enhance the grayscale value of the i-th pixel in the image, f ( i ) Let be the adjusted grayscale value of the i-th pixel, and let a and b be the minimum and maximum values ​​of the target grayscale values, respectively.

[0040] Finally, the feature judgment image obtained after grayscale adjustment is superimposed with the grayscale image to determine the location of the macroscopic uniformity feature in the door and window frame workpiece. Of course, the feature judgment image obtained here can also be superimposed with the previous frame image, and the macroscopic uniformity feature can also be located after superposition.

[0041] It should be noted that when door and window frame workpieces with uniformity issues are detected, their numbers and positioning results will be sent to the paint sprayer's terminal so that the paint sprayer can repair or adjust the problematic areas or locations.

[0042] Step 103: Apply a preset heat source excitation to the outer wall surface of the door and window frame workpiece that does not have macroscopic uniformity characteristics, and obtain a frame sequence thermal image by image matching of the thermal image obtained by the thermal imager and the frame image.

[0043] Currently, the manual method for inspecting uniformity ends after observing macroscopic uniformity characteristics such as missed paint spraying and bubbling; workpieces without these problems are then shipped out. However, it cannot be ruled out that workpieces without macroscopic problems may lack uniformity characteristics that are difficult to observe due to uneven paint thickness.

[0044] This application addresses the issue of uniformity features or problems that are difficult to observe by employing the following method: A preset heat source excitation is applied to the exterior wall surface of a door / window frame workpiece that lacks macroscopic uniformity features, and a corresponding thermal image is acquired using a thermal imager. Then, image matching is performed between the frame image and the thermal image to obtain a frame sequence thermal image of the door / window frame workpiece. It should be noted that the thermal image here is also a sequence image, arranged in a time series, recording the temperature distribution of the door / window frame workpiece under the preset heat source excitation, while the frame sequence thermal image only records the temperature distribution of each component of the frame.

[0045] In one possible implementation of this application, the aforementioned preset heat source excitation adopts a modulated heat source, whose output signal curve is a sine curve, and the output frequency is 0.1Hz and the power is 40W.

[0046] In one possible implementation of this application, obtaining a frame sequence thermal image of a door / window frame workpiece by image matching of a frame image and a thermal image includes: before applying a preset heat source excitation to the door / window frame workpiece, adjusting the image acquisition parameters of the thermal imager and the corresponding image acquisition parameters of the exterior wall image to be consistent. The image acquisition parameters include at least the image acquisition angle, image acquisition viewing angle, and image acquisition frame. That is, by adjusting the acquisition parameters of the image acquisition device and the thermal imager, it is ensured that the images acquired by the two are identical except for the content, providing a basis for subsequent image matching. Then, the frame image is marked in the exterior wall image, and the marked exterior wall image and the thermal image are superimposed to obtain a frame sequence thermal image that matches the frame image in the thermal image.

[0047] Step 104: Decompose the frame sequence heat map according to the polygonal shape of the door and window frame workpiece, and generate the temperature change curve of each component based on the decomposition result.

[0048] Furthermore, after obtaining the frame sequence heatmap, the frame sequence heatmap is further divided based on the polygonal shape corresponding to the door and window frame workpiece. For example, for a rectangular door and window frame workpiece, its corresponding frame sequence heatmap is divided into component sequence heatmaps corresponding to each rectangular side, resulting in four sets of component sequence heatmaps. For each set of component sequence heatmaps obtained from the division, its corresponding average temperature sequence is calculated, and temperature change curves are generated according to the average temperature sequence. Correspondingly, there are four temperature change curves, and when the preset heat source excitation is applied, the shape of the temperature change curve is the same as the shape of the preset heat source excitation. After the preset heat source excitation stops, the temperature change curve shows a downward trend. The temperature change curves here are used to characterize the correlation between the average temperature of each component of the door and window frame workpiece and time.

[0049] Figure 2Example 1 of the temperature change curve provided in this application represents a graph showing the excitation signal curve and temperature change curve under the excitation of a preset heat source. The horizontal axis represents time t, and the vertical axis represents temperature T, reflecting the change of temperature T with time t. Figure 2 There are two sine curves: the solid line represents the excitation signal curve generated by the preset heat source, while the dashed line represents the temperature change curve of a certain component. It can be seen that the temperature change curve and the excitation signal curve have the same shape, both being sine curves. It is important to note that there is a phase difference between the solid and dashed sine curves, which can be used to represent the thickness of the paint sprayed on the workpiece; when the thickness is consistent, the uniformity is also consistent. Therefore, under ideal uniformity conditions, the phase difference between the four temperature change curves obtained above and the excitation signal curve is the same.

[0050] Step 105: Obtain the temperature superposition curve based on the temperature change curve, and use the temperature superposition curve to determine the uniformity of paint spraying on the door and window frame workpiece.

[0051] Specifically, the time of the temperature change curves and excitation signal curves corresponding to each component of the door and window frame workpiece are aligned, and the temperature change curves are moved to the layer where the excitation signal curves are located to obtain the temperature superposition curves corresponding to the temperature change curves.

[0052] In one possible implementation of this application, after obtaining the temperature superposition curve, the uniformity of the paint spraying can be determined by the completeness of the superposition of the curve. Specifically, within the preset heat source excitation time period, the set of zero-crossing points in the temperature superposition curve is extracted. The elements in the set of zero-crossing points are arrays, and each array includes n data points, where n is the number of temperature change curves. That is, the first zero-crossing time corresponding to each of the four temperature change curves is the first array element in the set of zero-crossing points. For any array element in the zero-crossing point set, calculate its delay time relative to the corresponding zero-crossing point on the sine curve. For example, for the first array element in the set, calculate the time difference between the time data of each bit in the array and the time of the first zero-crossing point on the excitation signal curve. If the time difference fluctuates within the preset error range, the temperature superposition curves are determined to overlap into one curve. In this case, the paint uniformity of the door and window frame workpiece is considered to be high. However, if the time difference has a jump point relative to the preset error range, that is, if there are one or more zero-crossing point times in the zero-crossing point set that are significantly different from the zero-crossing point times on their corresponding sine curves, then the curve containing the jump zero-crossing point is identified as a jump curve. If there is at least one jump curve in the temperature superposition curve, the paint uniformity of the door and window frame workpiece is considered to be low.

[0053] Step 106: Traverse the frame sequence heatmaps corresponding to the non-uniform components through the rectangular test window, construct subsequence heatmaps, and use the temperature change curves generated by the subsequence heatmaps to determine the non-uniform location and degree of non-uniformity.

[0054] In one possible implementation of this application, after determining that there are components with low paint uniformity in the door and window frame workpiece, the components corresponding to the jump curve are identified as non-uniform components, and a rectangular test window is generated based on the size information of the non-uniform components. The width of the rectangular test window is equal to the width of the non-uniform components, and the length of the rectangular test window is a preset proportion of the length of the non-uniform components, so as to ensure that when traversing the frame sequence heatmap corresponding to the non-uniform components through the rectangular test window, an integer number of positions can be traversed.

[0055] Furthermore, in the frame sequence heatmap, regions corresponding to the same positions within the rectangular test window are extracted to construct sub-sequence heatmaps, i.e., heatmaps corresponding to a specific position. Temperature change curves corresponding to each sub-sequence heatmap are then generated, representing the temperature change at that position. Preferably, a portion of the heatmap after the preset heat source excitation stops is extracted from the frame sequence heatmap corresponding to the non-uniform component. This portion of the heatmap is then traversed through the rectangular test window to generate the temperature change curve of the non-uniform component during the cooling phase.

[0056] The temperature change curves corresponding to the subsequence heatmaps are time-aligned to identify separation curves among the relative temperature change curves. Based on the location of the rectangular test window corresponding to the separation curve, areas of uneven paint application are identified within the non-uniform component. It should be noted that the number of temperature change curves is equal to the number of rectangular test window locations on the non-uniform component; that is, each location corresponds to one temperature change curve. Ideally, these temperature change curves should have the same trend and the same temperature value at the same time, i.e., they should overlap. However, if one or more non-overlapping separate curves exist, it indicates the presence of one or more non-uniform locations or areas on the component.

[0057] Furthermore, after determining the separation curve, the starting separation time point corresponding to the separation curve is extracted. A straight line parallel to the time axis is used as the measurement line, and the temperature value corresponding to the measurement line is lower than the temperature value corresponding to the starting separation time point. If the measurement line passes through the separation curve first in the time axis direction, it indicates that the paint volume in the uneven painting area is low; conversely, if the measurement line passes through the temperature change curve first in the time axis direction and then through the separation curve, it indicates that the paint volume in the uneven painting area is high. This is because the higher the paint volume, the thicker the paint thickness in the uneven area, and the later the separation occurs on the temperature change curve.

[0058] Through the above process, it is possible to determine the components with uneven paint application, as well as the location or area of ​​unevenness on the components. It is also possible to determine whether the paint application is high or low at the location or area of ​​unevenness. This not only improves the accuracy and precision of uniformity detection, but also allows the detection results to be transmitted to the terminal, making it convenient for staff to repair or adjust the unevenness.

[0059] Based on the same inventive concept, this application also provides a device for detecting the uniformity of paint spraying during the production of door and window frames, the structure of which is as follows: Figure 3 As shown.

[0060] Figure 3 This application provides a structural schematic diagram of a device for testing the uniformity of paint application during the production of door and window frames. Figure 3 As shown, the paint uniformity testing device 300 in the door and window frame production process of this application specifically includes: at least one processor 301; and a memory 303 communicatively connected to at least one processor 301 (connected via bus 302); wherein, the memory 303 stores instructions that can be executed by at least one processor 301, so that at least one processor 301 can execute a paint uniformity testing method in the door and window frame production process as described in the above embodiments.

[0061] In one possible implementation of this application, the aforementioned processor is preferably an image processor, used to perform the following: acquiring an image of the exterior wall surface of a door / window frame workpiece after the painting process has ended; performing line detection on the acquired exterior wall surface image to crop the frame image corresponding to the door / window frame workpiece based on the line detection result; performing grayscale processing on the frame image to identify the macroscopic uniformity features of the door / window frame workpiece in the obtained grayscale image, the macroscopic uniformity features including at least one of missed paint, over-painting, peeling, and blistering; applying a preset heat source excitation to the exterior wall surface of the door / window frame workpiece that does not possess macroscopic uniformity features, and acquiring a thermal image of the door / window frame workpiece using a thermal imager; performing image matching on the thermal image using the frame image to obtain a frame sequence thermal image of the door / window frame workpiece; and using the polygonal shape corresponding to the door / window frame workpiece as a basis. The process involves splitting the frame sequence heatmap and calculating the average temperature sequence corresponding to each component sequence heatmap. Based on these average temperature sequences, temperature change curves are plotted for each component of the door / window frame workpiece. These curves characterize the relationship between the average temperature and time for each component. The time of the temperature change curves corresponding to each component of the door / window frame workpiece is aligned with the time of the corresponding signal curve from a preset heat source excitation. The temperature change curves are then moved to the layer containing the corresponding signal curve from the preset heat source excitation, resulting in a temperature overlay curve. If the temperature overlay curves overlap into a single curve, the paint uniformity of the door / window frame workpiece is determined to be high. If at least one skipped curve exists in the temperature overlay curve, the paint uniformity of the component corresponding to the skipped curve in the door / window frame workpiece is determined to be low.

[0062] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0063] The equipment and method provided in this application are one-to-one correspondences. Therefore, the equipment also has similar beneficial technical effects as its corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the equipment will not be repeated here.

[0064] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0065] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0066] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for detecting the uniformity of paint spraying during the production of door and window frames, characterized in that, The method includes: Image acquisition is performed on the exterior wall surface of the door and window frame workpiece after the painting process is completed, and line detection is performed on the acquired exterior wall surface image to obtain the frame image corresponding to the door and window frame workpiece based on the line detection result. The frame image is converted to grayscale to identify the macroscopic uniformity features of the door and window frame workpiece in the resulting grayscale image. The macroscopic uniformity features include at least one of the following: missed paint, over-painting, peeling, and blistering. A preset heat source excitation is applied to the outer wall surface of a door and window frame workpiece that does not possess the macroscopic uniformity characteristics, and a thermal image is acquired using a thermal imager. The thermal image is then matched with the frame image to obtain a frame sequence thermal image of the door and window frame workpiece. Based on the polygonal shape corresponding to the door and window frame workpiece, the frame sequence heat map is split into frames, and the average temperature sequence corresponding to each component sequence heat map obtained by splitting is calculated. Based on the average temperature sequence, the temperature change curves corresponding to each component of the door and window frame workpiece are plotted. The temperature change curves are used to characterize the correlation between the average temperature of each component of the door and window frame workpiece and time. Align the time of the temperature change curves corresponding to each component of the door and window frame workpiece with the time of the signal curve corresponding to the preset heat source excitation, and move the temperature change curve to the layer where the signal curve corresponding to the preset heat source excitation is located to obtain the temperature superposition curve corresponding to the temperature change curve. During the time period of the preset heat source excitation, the set of zero-crossing points in the temperature superposition curve is extracted. The elements in the set of zero-crossing points are arrays, and each array includes n data points, where n is the number of temperature change curves. The preset heat source excitation uses a modulated heat source, and its output signal curve is a sine curve with an output frequency of 0.1Hz. For any array element in the set of zero crossings, calculate its delay time relative to the corresponding zero crossing point on the sine curve; If the delay time fluctuates within the preset error range, it is determined that the temperature superposition curves overlap into one curve, and it is determined that the paint uniformity of the door and window frame workpiece is high. If the delay time jumps relative to the preset error range, the temperature change curve at the zero crossing point corresponding to the jump point is determined as the jump curve, and the paint uniformity of the component in the door and window frame workpiece corresponding to the jump curve is determined to be low.

2. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 1, characterized in that, The step of identifying the macroscopic uniformity features of the door and window frame workpiece in the obtained grayscale image includes: Based on the size information of each component in the door and window frame workpiece, a difference calculation window is generated in the grayscale image, and the length of the difference calculation window is a preset ratio of the length of each component; The grayscale image is traversed using the difference calculation window, and the previous window region and the next window region corresponding to the difference calculation window are subtracted backward. The original grayscale value in the previous window region is replaced with the value after difference calculation to obtain the enhanced image of the grayscale image. Calculate the gray-level histogram corresponding to the enhanced image, and determine the target gray-level value of the enhanced image in the gray-level histogram; The enhanced image is adjusted using the target grayscale value to obtain a feature judgment image, in order to identify the macroscopic uniformity features of the door and window frame workpiece in the feature judgment image.

3. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 2, characterized in that, After determining the target grayscale value of the enhanced image from the grayscale histogram, the method further includes: If the target grayscale value is zero, then it is determined that the macroscopic uniformity feature does not exist in the door and window frame workpiece; If the target grayscale value is not zero, the enhanced image is adjusted for grayscale using the following formula: in, Let be the grayscale value of the i-th pixel in the enhanced image. Let be the adjusted grayscale value of the i-th pixel, and a and b be the minimum and maximum values ​​among the target grayscale values, respectively. The feature judgment map obtained after grayscale adjustment is superimposed with the grayscale image and / or the frame image to achieve feature localization of the macroscopic uniformity feature.

4. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 1, characterized in that, After determining that the paint uniformity of the component corresponding to the jump curve in the door and window frame workpiece is low, the method further includes: The component corresponding to the jump curve is identified as a non-uniform component, and a rectangular test window is generated based on the size information of the non-uniform component. The width of the rectangular test window is equal to the width of the non-uniform component, and the length of the rectangular test window is a preset proportion of the length of the non-uniform component. The frame sequence heatmap corresponding to the non-uniform component is traversed through the rectangular test window. In the frame sequence heatmap, the regions with the same position corresponding to the rectangular test window are extracted to construct subsequence heatmaps, and temperature change curves corresponding to the subsequence heatmaps are generated respectively; The temperature change curves corresponding to the heatmaps of the subsequences are time-aligned to determine the separation curves relative to the temperature change curves in the time-aligned curves. Based on the location of the rectangular test window corresponding to the separation curve, the uneven paint spraying area is determined in the uneven component.

5. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 4, characterized in that, After determining the separation curve relative to the temperature change curve, the method further includes: Extract the start separation time node corresponding to the separation curve from the time-aligned curve graph; A straight line parallel to the time axis is used as the measurement line, and the temperature value corresponding to the measurement line is lower than the temperature value corresponding to the start of the separation time node; If the measurement line passes through the separation curve first in the time axis direction, it is determined that the amount of paint sprayed in the uneven paint spraying area is low. If the measurement line passes through the temperature change curve first and then the separation curve in the time axis direction, then it is determined that the paint volume in the uneven paint area is high.

6. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 4, characterized in that, Traversing the frame sequence heatmap corresponding to the non-uniform component through the rectangular test window includes: Extract a portion of the thermal image after the preset heat source excitation stops from the frame sequence thermal image corresponding to the non-uniform component. Traverse the portion of the thermal image through the rectangular test window to generate the temperature change curve of the non-uniform component during the cooling phase.

7. The method for detecting the uniformity of paint spraying during the production of door and window frames according to claim 1, characterized in that, The frame sequence thermal image of the door and window frame workpiece is obtained by image matching of the thermal image with the frame image, including: Before applying a preset heat source excitation to the door and window frame workpiece, the image acquisition parameters of the thermal imager and the image acquisition parameters corresponding to the exterior wall image are adjusted to be consistent. The image acquisition parameters include at least the image acquisition angle, the image acquisition viewing angle, and the image acquisition frame. The frame image is marked in the exterior wall image, and the marked exterior wall image and the thermal image are superimposed to obtain a frame sequence thermal image that matches the frame image in the thermal image.

8. A device for testing the uniformity of paint spraying during the production of door and window frames, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method for detecting the uniformity of paint spraying during the production of door and window frames according to any one of claims 1-7.

Images