Dry powder removal rate detection method and device, equipment and medium

Abstract

The invention provides a dry powder removal rate detection method and device, equipment and a medium, and belongs to the technical field of metallurgical automatic detection.The method comprises the steps that first image data of sintered particles are processed to obtain a first powder removal rate, and the first image data are RGB images of the sintered particles; a three-dimensional model of the sintered particles is obtained according to second image data of the sintered particles, a second powder removal rate is obtained through calculation according to the three-dimensional model, and the second image data is a multi-angle image of the sintered particles; third image data of the sintered particles are analyzed, a third powder removal rate is obtained, and the third image data are continuous frame images of the sintered particles; and performing weighted calculation on the first powder removal rate, the second powder removal rate and the third powder removal rate to obtain a target powder removal rate. According to the dry powder removal rate detection method and device, the equipment and the medium provided by the invention, comprehensive analysis is carried out by adopting multi-source image data, and a comprehensive and accurate detection result is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of metallurgical automation detection, and more specifically, relates to a drying powder loss rate detection method, device, equipment and medium. Background Art

[0002] Sintering processes are widely used in the material production field, and the quality of sintered pellets directly impacts product performance. As a key indicator of sintered pellet quality, the drying powder loss rate is crucial for accurate measurement. However, existing measurement methods have many limitations. For example, simple weighing methods can only provide a rough estimate of the amount of powder loss, but cannot fully reflect the surface condition and internal structure of the pellets, as well as the dynamic changes in powder loss during the drying process.

[0003] In a complex industrial production environment, different batches of sintered particles are affected by differences in raw materials and fluctuations in the sintering process. The powder loss is complex and changeable, and the errors of conventional detection methods continue to accumulate, which cannot meet the production needs for high-precision quality control.

[0004] Therefore, there is an urgent need for a method, device, equipment and medium for detecting the drying powder loss rate. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and device for detecting the drying powder loss rate, so as to improve the accuracy of the detection.

[0006] A first aspect of an embodiment of the present invention provides a method for detecting a drying powder loss rate, comprising: Processing first image data of the sintered particles to obtain a first powder removal rate, wherein the first image data is an RGB image of the sintered particles; obtaining a three-dimensional model of the sintered particles according to second image data of the sintered particles, and calculating a second powder removal rate according to the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles; analyzing third image data of the sintered particles to obtain a third powder removal rate, wherein the third image data is a continuous frame image of the sintered particles; A target powder removal rate is obtained by performing weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate.

[0007] A second aspect of the embodiments of the present invention provides a drying powder loss rate detection device, comprising: a first data processing module, configured to process first image data of the sintered particles to obtain a first powder removal rate, wherein the first image data is an RGB image of the sintered particles; a second data processing module, configured to obtain a three-dimensional model of the sintered particles based on second image data of the sintered particles, and calculate a second powder removal rate based on the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles; a third data processing module, configured to analyze third image data of the sintered particles to obtain a third powder removal rate, wherein the third image data is a continuous frame image of the sintered particles; The fourth data processing module is used to perform weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate to obtain a target powder removal rate.

[0008] According to a third aspect of an embodiment of the present invention, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned method for detecting a drying powder loss rate when executing the computer program.

[0009] According to a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for detecting a drying powder loss rate are implemented.

[0010] The beneficial effects of a dry powder removal rate detection method and device, electronic device, and readable storage medium provided by an embodiment of the present invention are as follows: This embodiment first obtains a first powder removal rate by processing the RGB images of sintered particles, which can reflect the powder removal status from the intuitive dimensions of surface color and texture, and quickly locate the surface powder removal area and proportion. Secondly, this embodiment considers the impact of powder removal in depth from the perspective of volume and shape changes, and uses multi-angle images to construct a three-dimensional model to calculate the second powder removal rate. Finally, this embodiment can analyze continuous frame images to obtain a third powder removal rate, and capture powder removal caused by collision and friction in a dynamic process. In summary, the present invention comprehensively calculates the powder removal rate through multi-dimensional data, improves the accuracy of detection, and comprehensively presents the true level of powder removal of sintered particles. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 A schematic flow chart of a method for detecting a drying powder loss rate according to an embodiment of the present invention; Figure 2 This is a structural block diagram of a drying powder loss rate detection device provided by one embodiment of the present invention; Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0014] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below with reference to the accompanying drawings.

[0015] Please refer to Figure 1 , Figure 1 A schematic flow chart of a method for detecting a drying powder loss rate according to an embodiment of the present invention is provided, wherein the method comprises: S101: Processing first image data of sintered particles to obtain a first powder removal rate, where the first image data is an RGB image of the sintered particles.

[0016] In this embodiment, first image data of sintered particles is processed to obtain a first powder loss rate. The first image data is an RGB image of the sintered particles. In actual industrial production scenarios, sintered particles will experience surface powder loss after undergoing a series of processes, such as drying. RGB images can intuitively display visual characteristics of sintered particles, such as color and texture. By processing this image data, the degree of powder loss can be analyzed. This embodiment uses a high-resolution RGB camera to capture the sintered stone particles, enabling the acquisition of clear first image data.

[0017] For example, processing the first image data using an image recognition algorithm includes grayscaling the image to convert the RGB color space into grayscale space, simplifying subsequent calculations. Next, an edge detection algorithm, such as the Canny algorithm, is used to identify the outlines of sintered particles and any de-powdered particles. The first de-powdering rate is calculated by calculating the ratio of the number of pixels per de-powdered particle to the total number of pixels per sintered particle and de-powdered particle. For example, if, in a processed image, the number of pixels per de-powdered particle is 1,000 and the total number of pixels per sintered particle and de-powdered particle is 10,000, then the first de-powdering rate is 1000 ÷ 10,000 × 100% = 10%.

[0018] S102: obtaining a three-dimensional model of the sintered particles according to second image data of the sintered particles, and calculating a second powder removal rate according to the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles.

[0019] In this embodiment, a three-dimensional model of the sintered particles is obtained according to the second image data of the sintered particles, and the second powder removal rate is calculated according to the three-dimensional model. The second image data is a multi-angle image of the sintered particles.

[0020] In this embodiment, a camera captures the sintered particles from multiple angles to obtain second image data. The uniform distribution of the shooting angles allows for comprehensive capture of the shape characteristics of the sintered particles. The multi-angle images provide information about the sintered particles from different angles. By processing these multi-angle images, such as through stereo matching and 3D reconstruction techniques, a 3D model of the sintered particles can be constructed. This 3D model comprehensively reflects information such as the shape and volume of the sintered particles. By comparing this model with an ideal model without de-powdering, the volume loss due to de-powdering can be calculated, thereby obtaining a second de-powdering rate.

[0021] S103: Analyze third image data of the sintered particles to obtain a third powder removal rate, where the third image data is a continuous frame image of the sintered particles.

[0022] In this embodiment, a high-speed camera is used to continuously shoot the state of the sintered particles during the drying process to obtain third image data, which are continuous frame images of the sintered particles; the continuous frame images record the dynamic changes of the sintered particles over a period of time. During this process, the sintered particles may collide, rub, and so on, which may cause powder shedding. By analyzing continuous frame images, such as using algorithms such as inter-frame difference method, target tracking algorithm, optical flow method, etc., tracking the motion trajectory of the particles in the image, etc., the dynamic process of powder shedding on the surface of the sintered particles can be identified. For example, it is observed that new powder shedding areas appear on the surface of the sintered particles after the collision. By analyzing and counting these powder shedding areas, the third powder shedding rate can be calculated.

[0023] S104: performing weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate to obtain a target powder removal rate.

[0024] In this embodiment, the first powder removal rate, the second powder removal rate, and the third powder removal rate are weighted to obtain a target powder removal rate. Different powder removal rate calculation methods reflect the powder removal situation of sintered particles from different perspectives, and each method has its advantages and disadvantages. According to actual needs and the reliability of different powder removal rate detection methods, weights are set for the first powder removal rate, the second powder removal rate, and the third powder removal rate. Through weighted calculation, a more accurate and comprehensive target powder removal rate can be obtained. For example, if it is believed that the first powder removal rate detection based on the surface image can better reflect the current surface powder removal situation, its weight can be set to 0.4; the second powder removal rate detection based on the three-dimensional model can reflect the influence of the internal structure on the powder removal, and the weight is set to 0.3; the third powder removal rate detection based on the continuous frame image can reflect the dynamic process, and the weight is set to 0.3.

[0025] From the above, it can be concluded that this embodiment first obtains the first de-powdering rate by processing the RGB images of the sintered particles, which can reflect the de-powdering status from the intuitive dimensions of surface color and texture, and quickly locate the surface de-powdering area and proportion. Secondly, this embodiment considers the impact of de-powdering in depth from the perspective of volume and shape changes, and uses multi-angle images to construct a three-dimensional model to calculate the second de-powdering rate. Finally, this embodiment can analyze continuous frame images to obtain the third de-powdering rate, and capture the de-powdering caused by collision and friction in the dynamic process. In summary, the present invention comprehensively calculates the de-powdering rate through multi-dimensional data, improves the accuracy of detection, and comprehensively presents the true level of de-powdering of sintered particles.

[0026] In one embodiment of the present invention, processing first image data of sintered particles to obtain a first powder removal rate includes: The RGB image of the sintered particles is preprocessed to obtain an HSV image; The HSV image is processed using an image segmentation algorithm to obtain the powder-free areas in each channel of the HSV image; The first powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles.

[0027] In this example, an RGB image of sintered particles is preprocessed to generate an HSV image. While the RGB color space primarily represents color using the red, green, and blue channels, the HSV color space describes color using hue, saturation, and value. In practical applications, the HSV color space is more intuitive and consistent with human visual perception. Converting RGB images to HSV images facilitates color analysis and processing.

[0028] This example uses an image segmentation algorithm to process the HSV image and obtain the de-powdered regions in each channel of the HSV image. Because the de-powdered regions and the main sintered particle regions typically have different color and brightness characteristics in each channel of the HSV image (H channel, S channel, and V channel), by properly setting the segmentation parameters, the de-powdered regions can be accurately separated from the HSV image, obtaining the de-powdered regions in each channel of the HSV image. Common image segmentation algorithms include threshold segmentation, edge detection, and region growing.

[0029] In this embodiment, after obtaining the de-powdering area in each channel of the HSV image, it is necessary to determine the reference surface area of the sintered particles. By performing edge detection on the original RGB image, the outline of the sintered particles is accurately identified, and the reference surface area of the sintered particles is calculated. The area of the de-powdering area in each channel of the HSV image is calculated separately (this can be done by counting the number of pixels in the de-powdering area and then converting it into the actual area based on the resolution of the image). The area of the de-powdering area in each channel is added together to obtain the total area of the de-powdering area. The first de-powdering rate is calculated based on the ratio of the total de-powdering area area to the reference surface area of the sintered particles. For example, if the area of the de-powdering area is 10 square centimeters and the reference surface area of the sintered particles is 100 square centimeters, the first de-powdering rate is 10%.

[0030]

[0031] in, is the first powder removal rate calculated based on the HSV image; is the channel weight , ; is the area of the powder removal region in each channel (number of pixels or actual physical area), which is extracted in each HSV channel by the image segmentation algorithm; is the reference surface area of the sintered particles (number of pixels or actual physical area), and is the total surface area of the intact sintered particles without de-powdering.

[0032] In one embodiment of the present invention, processing first image data of sintered particles to obtain a first powder removal rate includes: Preprocess the RGB image of the sintered particles to obtain the powder-depleted areas in each channel of the RGB image; The fourth powder removal rate is calculated based on the ratio of the powder removal area in each channel of the RGB image to the reference surface area of the sintered particles; The RGB image of the sintered particles is preprocessed to obtain an HSV image; The HSV image is processed using an image segmentation algorithm to obtain the powder-free areas in each channel of the HSV image; The fifth powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles; The first powder removal rate is obtained by weighted calculation of the fourth powder removal rate and the fifth powder removal rate.

[0033] In this embodiment, an RGB image of sintered particles is preprocessed to determine the de-powdered regions within each channel of the RGB image. Preprocessing includes image filtering, enhancement, and other operations to improve image quality and clarity. This embodiment uses color feature analysis, combined with the color differences between sintered particles and de-powdered particles in the three RGB channels (red channel R, green channel G, and blue channel B), and sets appropriate threshold ranges to determine the de-powdered regions within each channel of the RGB image. The number of pixels in the de-powdered regions within each channel of the RGB image is then counted, and the total number of pixels in the de-powdered regions is summed across the three channels to obtain the total number of pixels in the de-powdered regions. Simultaneously, the number of pixels within the entire sintered particle is counted (this can be accomplished by performing edge detection on the image to determine the particle outline and then counting the pixels within the outline). A fourth de-powdering rate is calculated based on the ratio of the de-powdered regions within each channel of the RGB image to the baseline surface area of the sintered particle.

[0034] The fourth powder removal rate calculation formula is:

[0035] in, The fourth powder removal rate; is the RGB image channel weight, ; is the de-powdering area of each channel (number of pixels or actual physical area), which is obtained by image segmentation algorithm. Extraction in each channel; is the reference surface area of the sintered particles (number of pixels or actual physical area), and is the total surface area of the intact sintered particles without de-powdering; It is the RGB channel brightness correction function; The calculation formula of the RGB channel brightness correction function is:

[0036] in, is the channel brightness mean, are the global brightness mean and standard deviation of the training set, is the sensitivity coefficient (default 0.1); The powder removal area weights of the RGB channels can be dynamically modified according to the lighting conditions.

[0037] In this embodiment, an RGB image of sintered particles is preprocessed to obtain an HSV image. The HSV image is processed using an image segmentation algorithm to obtain the de-powdering regions in each channel of the HSV image. A fifth de-powdering rate is calculated based on the ratio of the de-powdering regions in each channel of the HSV image to the reference surface area of the sintered particles. The fourth and fifth de-powdering rates are weighted to obtain a first de-powdering rate. The fourth and fifth de-powdering rates reflect the de-powdering of the sintered particles in the RGB color space and the HSV color space, respectively. Because different color spaces differ in expressing color and identifying de-powdering regions, a more accurate first de-powdering rate can be obtained by combining the results of these two methods through weighted calculation.

[0038] In this embodiment, the fifth powder removal rate calculation formula is:

[0039] in, is the fifth powder removal rate based on HSV image; is the HSV channel saturation enhancement factor; The calculation formula is:

[0040] in, is the reference saturation (such as the average value of normal particles), To control the decay rate. It is the statistical value of the saturation (S channel) of the HSV color space in the current image; The first powder removal rate calculation formula is:

[0041] in, is the first powder removal rate, The fourth powder removal rate, is the fifth powder removal rate based on HSV image; is the first weight, is the second weight. The first powder loss rate calculation formula in this embodiment is more accurate than the first powder loss rate calculation formula in the previous embodiment. The fourth and fifth powder loss rates reflect the powder loss of sintered particles using the RGB color space and HSV color space, respectively. Because different color spaces differ in expressing color and identifying powder loss areas, a weighted calculation combines the results of these two methods to obtain a more accurate first powder loss rate.

[0042] In one embodiment of the present invention, the first weight adjustment step size is determined based on the ambient light intensity; Adjusting the first weight reference value based on the first weight adjustment step to obtain a first weight; adjusting the second weight reference value based on the first weight adjustment step to obtain a second weight; A first fan removal rate is obtained by performing weighted calculation based on the first weight, the second weight, the fourth fan removal rate, and the fifth fan removal rate; The first weight is a weight corresponding to the fourth powder removal rate, the second weight is a weight corresponding to the fifth powder removal rate, and the adjustment directions of the first weight reference value and the second weight reference value are different.

[0043] In this embodiment, the ambient light intensity will affect the quality and color information of the image. Under different light intensities, the recognition accuracy of the de-powdered areas in RGB images and HSV images will be different. For example, under strong light, the image will be overexposed, resulting in unclear color features of the de-powdered areas; under weak light, the image will be blurred, which will also affect the recognition of the de-powdered areas. Therefore, it is necessary to adjust the weights of the fourth de-powdering rate and the fifth de-powdering rate according to the ambient light intensity. The first weight adjustment step indicates the magnitude of the weight adjustment under different light intensities. The relationship between light intensity and weight adjustment step can be established through experimental data. For example, when the light intensity is strong, the weight adjustment step is appropriately increased to adjust the weight more flexibly.

[0044] In this embodiment, the calculation formula for the weight adjustment step is:

[0045] in, is the first weight adjustment step size, is the reference light intensity; are the upper and lower limits of light intensity respectively; To adjust the step size coefficient.

[0046] The first weight reference value is adjusted based on the first weight adjustment step to obtain the first weight; the second weight reference value is adjusted based on the first weight adjustment step to obtain the second weight. The first weight reference value and the second weight reference value are pre-set initial weight values, corresponding to the fourth de-powdering rate and the fifth de-powdering rate, respectively. According to the first weight adjustment step determined by the ambient light intensity, these two reference values are adjusted to obtain the final first weight and second weight. For example, if the first weight reference value is 0.6 and the second weight reference value is 0.4, under the current light intensity, the first weight adjustment step is 0.1, and the first weight needs to be increased, then the first weight is adjusted to 0.7 and the second weight is adjusted to 0.3.

[0047] The first powder removal rate is calculated using a weighted combination of the first weight, the second weight, the fourth powder removal rate, and the fifth powder removal rate. The first weight corresponds to the fourth powder removal rate, and the second weight corresponds to the fifth powder removal rate. The first and second weight reference values are adjusted in different directions. When adjusting the weights, the first and second weights are adjusted in opposite directions to ensure that their sum is always 1. This method allows the contributions of the fourth and fifth powder removal rates to the calculation of the first powder removal rate to be dynamically adjusted based on ambient light intensity, thereby improving the accuracy of the first powder removal rate calculation.

[0048] In one embodiment of the present invention, obtaining a three-dimensional model of the sintered particles according to the second image data of the sintered particles, and calculating the second powder removal rate according to the three-dimensional model include: Processing multi-angle images to obtain three-dimensional point cloud data of sintered particles; A three-dimensional model of sintered particles is established based on the three-dimensional point cloud data, and the volume of the three-dimensional model is calculated based on the three-dimensional model; The volume loss rate of the sintered particles is obtained by comparing the volume of the three-dimensional model with the reference model without de-powdering, and the volume loss rate is used as the second de-powdering rate.

[0049] In this embodiment, the multi-angle image provides information of sintered particles at different angles. Before processing the multi-angle image, the collected multi-angle image needs to be preprocessed. The preprocessing operations include image denoising, grayscale conversion, and image enhancement. Feature points are extracted from the preprocessed image; matching feature point pairs are found between images at different angles through a feature matching algorithm. Based on the matching feature point pairs, the coordinates of each feature point in three-dimensional space are calculated, and all the calculated three-dimensional coordinate points are combined to obtain three-dimensional point cloud data of the sintered particles; the three-dimensional point cloud data represents the surface shape and spatial position information of the sintered particles in the form of discrete points. The three-dimensional point cloud data is processed and reconstructed to obtain a three-dimensional model of the sintered particles; the volume loss rate of the sintered particles is obtained by comparing the volume of the three-dimensional model with the un-de-powdered reference model, and the volume loss rate is used as the second de-powdering rate.

[0050] In this embodiment, the unde-powdered reference model refers to an ideal three-dimensional model of the sintered particle before de-powdering occurs. Its volume is established using the same image acquisition and three-dimensional model reconstruction methods as the unde-powdered reference model. The volume of the unde-powdered reference model represents the original volume of the sintered particle. This embodiment compares the volume of the current sintered particle's three-dimensional model with the volume of the unde-powdered reference model to calculate the ratio of volume loss, i.e., the volume loss rate. The volume loss rate reflects the reduction in the sintered particle volume due to de-powdering and is used as the second de-powdering rate.

[0051] In one embodiment of the present invention, processing multi-angle images to obtain three-dimensional point cloud data of sintered particles includes: Preprocessing the multi-angle images to obtain first point cloud data; determining a voxel size according to the point cloud density of the first point cloud data; The first point cloud data is registered based on a normal distribution transformation algorithm and voxel size to obtain three-dimensional point cloud data.

[0052] In this embodiment, preprocessing includes operations such as image filtering and feature extraction. By preprocessing multi-angle images, this embodiment can extract feature points from the images and calculate three-dimensional coordinates based on the matching relationships between these feature points, thereby obtaining first point cloud data. For example, the Scale-Invariant Feature Transform (SIFT) algorithm can be used to extract feature points from the images, and then the three-dimensional coordinates of these feature points can be calculated using a stereo matching algorithm.

[0053] In this embodiment, point cloud density refers to the number of points per unit volume and reflects the sparsity of the point cloud data. A voxel is a small cube in three-dimensional space, and its size determines the accuracy of the three-dimensional space division. When determining the voxel size, the point cloud density should be considered. If the point cloud density is high, a smaller voxel size can be selected to improve the accuracy of the 3D model; if the point cloud density is low, a larger voxel size should be selected to avoid situations where no points are present in the voxel.

[0054] In this embodiment, the first point cloud data is registered based on the normal distribution transformation algorithm and voxel size to obtain three-dimensional point cloud data. The purpose of the registration is to align the point cloud data obtained from different perspectives so that they are in the same coordinate system. The point cloud data is modeled as a probability density function of the normal distribution through the normal distribution transformation algorithm, and the optimal transformation parameters are found so that the point cloud data from different perspectives can be accurately aligned; in the registration process, based on the determined voxel size, the statistical characteristics of the point cloud data in each voxel, such as the mean and covariance matrix, are calculated; the optimal rotation and translation parameters are solved through iterative optimization, and the point cloud data from different perspectives are unified into the same coordinate system to obtain three-dimensional point cloud data. The three-dimensional point cloud data obtained according to the normal distribution transformation algorithm accurately reflects the three-dimensional shape and structural information of the sintered particles.

[0055] This embodiment can also use point cloud processing software (such as a point cloud library) to process and reconstruct the three-dimensional point cloud data to obtain a three-dimensional model of sintered particles; and use a voxelization method or an integration-based method to calculate the volume of the three-dimensional model of sintered particles.

[0056] In one embodiment of the present invention, analyzing the third image data of the sintered particles to obtain a third powder removal rate includes: The motion trajectory of the sintered particles is extracted by the inter-frame difference method; the trajectory curvature and motion trajectory are analyzed and identified to obtain the key frame position where the collision event occurs; Based on the keyframe position, sub-pixel registration is performed on the sintered particle ROI area within the time sequence window before and after the collision event; Statistically calculate the grayscale reduction of the surface pixel grayscale values of the sintered particles before and after the collision of the sintered particles; When the grayscale reduction is greater than the set threshold, it is determined to be a powder-shedding area; and the third powder-shedding rate is calculated based on the ratio of the area of the powder-shedding area to the reference surface area of the sintered particles.

[0057] In this embodiment, the motion trajectory of sintered particles is extracted using an inter-frame difference method. Specifically, adjacent frame images are differentiated, and the difference results are subjected to threshold processing and morphological operations to extract the motion region of the sintered particles and track their motion trajectory. This embodiment uses trajectory curvature analysis to identify the keyframe locations where collision events occur. Curvature analysis is performed on the motion trajectory of the sintered particles, calculating the curvature value of each point on the trajectory. When the curvature value exceeds a certain threshold, a collision event is considered to have occurred, and the corresponding frame is designated as a keyframe.

[0058] Sub-pixel registration is performed on the ROI (Region of Interest) of the sintered particles within the timing window before and after the collision event (keyframe). After the collision event, powder will be removed from the surface of the sintered particles, resulting in a decrease in the grayscale value of the surface pixels. The magnitude of the grayscale decrease is calculated by comparing the grayscale values of the surface pixels of the sintered particles before and after the collision.

[0059] When the grayscale decrease exceeds a set threshold, it is determined to be a powder loss area. The threshold is determined based on experimental data and actual application scenarios, and is used to determine whether the grayscale decrease is caused by powder loss. If the grayscale decrease in a region exceeds the set threshold, the region is considered to have experienced powder loss.

[0060] The third powder removal rate is calculated based on the ratio of the powder removal area to the reference surface area of the sintered particles. This can be done by counting the number of pixels within the powder removal area and then converting the number of pixels into the actual area based on the image resolution.

[0061] In this embodiment, the third powder removal rate calculation formula is:

[0062] in, The third powder removal rate is the percentage of powder removal area due to collision events; is the area of the powder-shedding region detected in the J-th collision event; is the grayscale attenuation confidence function, which is proportional to the grayscale reduction amplitude Related.

[0063] The present invention effectively detects surface shedding caused by mechanical impact by capturing instantaneous grayscale changes during the collision process; it complements static image analysis and significantly improves the ability of powder shedding detection.

[0064] Corresponding to the drying powder loss rate detection method of the above embodiment, Figure 2 This is a structural block diagram of a device for detecting a drying powder loss rate according to an embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown. Figure 2 The drying powder removal rate detection device 20 includes: a first data processing module 21, a second data processing module 22, a third data processing module 23 and a fourth data processing module 24.

[0065] The first data processing module 21 is used to process the first image data of the sintered particles to obtain a first powder removal rate, where the first image data is an RGB image of the sintered particles; A second data processing module 22 is configured to obtain a three-dimensional model of the sintered particles based on the second image data of the sintered particles, and calculate a second powder removal rate based on the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles; A third data processing module 23 is configured to analyze third image data of the sintered particles to obtain a third powder removal rate, where the third image data is a continuous frame image of the sintered particles; The fourth data processing module 24 is configured to perform weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate to obtain a target powder removal rate.

[0066] In one embodiment of the present invention, the first data processing module 21 is specifically configured to: The RGB image of the sintered particles is preprocessed to obtain an HSV image; The HSV image is processed using an image segmentation algorithm to obtain the powder-free areas in each channel of the HSV image; The first powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles.

[0067] In one embodiment of the present invention, the first data processing module 21 is specifically configured to: Preprocess the RGB image of the sintered particles to obtain the powder-depleted areas in each channel of the RGB image; The fourth powder removal rate is calculated based on the proportion of the powder removal area in each channel of the RGB image to the entire sintered particles; The RGB image of the sintered particles is preprocessed to obtain an HSV image; The HSV image is processed using an image segmentation algorithm to obtain the powder-free areas in each channel of the HSV image; The fifth powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles; The first powder removal rate is obtained by weighted calculation of the fourth powder removal rate and the fifth powder removal rate.

[0068] In one embodiment of the present invention, the first data processing module 21 is specifically configured to: determining a first weight adjustment step size based on ambient light intensity; Adjusting the first weight reference value based on the first weight adjustment step to obtain a first weight; adjusting the second weight reference value based on the first weight adjustment step to obtain a second weight; A first fan removal rate is obtained by performing weighted calculation based on the first weight, the second weight, the fourth fan removal rate, and the fifth fan removal rate; The first weight is a weight corresponding to the fourth powder removal rate, the second weight is a weight corresponding to the fifth powder removal rate, and the adjustment directions of the first weight reference value and the second weight reference value are different.

[0069] In one embodiment of the present invention, the second data processing module 22 is specifically configured to: Processing multi-angle images to obtain three-dimensional point cloud data of sintered particles; A three-dimensional model of sintered particles is established based on the three-dimensional point cloud data, and the volume of the three-dimensional model is calculated based on the three-dimensional model; The volume loss rate of the sintered particles is obtained by comparing the volume of the three-dimensional model with the reference model without de-powdering, and the volume loss rate is used as the second de-powdering rate.

[0070] In one embodiment of the present invention, the second data processing module 22 is specifically configured to: Preprocessing the second image data to obtain first point cloud data; determining a voxel size according to the point cloud density of the first point cloud data; The first point cloud data is registered based on a normal distribution transformation algorithm and voxel size to obtain three-dimensional point cloud data.

[0071] In one embodiment of the present invention, the second data processing module 23 is specifically configured to: The motion trajectory of sintered particles is extracted by using the inter-frame difference method; Using trajectory curvature and motion trajectory analysis to identify the key frame locations where collision events occur; Perform sub-pixel registration of the sintered particle ROI area within the time sequence window before and after the collision event; Statistically calculate the grayscale reduction of the surface pixel grayscale values before and after the sintered particles collide; When the grayscale decrease is greater than the set threshold, it is determined as a powder-depleted area; The third powder removal rate is calculated based on the ratio of the powder removal area to the reference surface area of the sintered particles.

[0072] This embodiment first obtains a first de-powdering rate by processing the RGB images of the sintered particles, which can reflect the de-powdering status from the intuitive dimensions of surface color and texture, and quickly locate the surface de-powdering area and proportion. Secondly, this embodiment considers the impact of de-powdering from the perspective of volume and shape changes in depth, and uses multi-angle images to construct a three-dimensional model to calculate the second de-powdering rate. Finally, this embodiment can analyze continuous frame images to obtain a third de-powdering rate, capturing the de-powdering caused by collision and friction during the dynamic process. In summary, the present invention comprehensively calculates the de-powdering rate through multi-dimensional data, improves the accuracy of detection, and comprehensively presents the true level of de-powdering of sintered particles.

[0073] See also Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided by an embodiment of the present invention. Figure 3 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 2 The functions of the first data processing module 21, the second data processing module 22, the third data processing module 23 and the fourth data processing module 24 are shown.

[0074] It should be understood that in the embodiment of the present invention, the processor 301 may be a central processing unit (CPU), and may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0075] The input device 302 may include a touchpad, a fingerprint collection sensor (for collecting user fingerprint information and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0076] The memory 304 may include a read-only memory and a random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store device type information.

[0077] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiment of the present invention can execute the implementation methods described in the first and second embodiments of the drying powder removal rate detection method provided in the embodiment of the present invention, and can also execute the implementation method of the electronic device described in the embodiment of the present invention, which will not be repeated here.

[0078] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. The computer program includes program instructions. When executed by a processor, the program instructions implement all or part of the process of the method in the above-mentioned embodiment. The computer program can also be used to instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. Computer-readable media can include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium.

[0079] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the computer-readable storage medium can include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or is about to be output.

[0080] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0081] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0082] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or can be an electrical, mechanical or other form of connection.

[0083] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of the embodiments of the present invention.

[0084] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0085] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for detecting drying powder loss rate, characterized in that: include: Processing first image data of the sintered particles to obtain a first powder removal rate, wherein the first image data is an RGB image of the sintered particles; obtaining a three-dimensional model of the sintered particles according to second image data of the sintered particles, and calculating a second powder removal rate according to the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles; analyzing third image data of the sintered particles to obtain a third powder removal rate, wherein the third image data is a continuous frame image of the sintered particles; A target powder removal rate is obtained by performing weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate.

2. The method for detecting the drying powder loss rate according to claim 1, wherein: The processing of the first image data of the sintered particles to obtain the first powder removal rate includes: Preprocessing the RGB image of the sintered particles to obtain an HSV image; Processing the HSV image using an image segmentation algorithm to obtain powder-free areas in each channel of the HSV image; The first powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles.

3. The method for detecting the drying powder loss rate according to claim 1, wherein: The processing of the first image data of the sintered particles to obtain a first powder removal rate includes: Preprocessing the RGB image of the sintered particles to obtain powder-depleted areas in each channel of the RGB image; The fourth powder removal rate is calculated based on the ratio of the powder removal area in each channel of the RGB image to the reference surface area of the sintered particles; Preprocessing the RGB image of the sintered particles to obtain an HSV image; Processing the HSV image using an image segmentation algorithm to obtain powder-free areas in each channel of the HSV image; The fifth powder removal rate is calculated based on the ratio of the powder removal area in each channel of the HSV image to the reference surface area of the sintered particles; The first powder removal rate is obtained by performing weighted calculation on the fourth powder removal rate and the fifth powder removal rate.

4. The method for detecting the drying powder loss rate according to claim 3, wherein: Also includes: determining a first weight adjustment step size based on ambient light intensity; Adjusting a first weight reference value based on the first weight adjustment step to obtain a first weight; Adjusting a second weight reference value based on the first weight adjustment step to obtain a second weight; A first powder removal rate is obtained by performing weighted calculation based on the first weight, the second weight, the fourth powder removal rate, and the fifth powder removal rate; The first weight is a weight corresponding to the fourth powder removal rate, the second weight is a weight corresponding to the fifth powder removal rate, and the adjustment directions of the first weight reference value and the second weight reference value are different.

5. The method for detecting the drying powder loss rate according to claim 1, wherein: The step of obtaining a three-dimensional model of the sintered particles according to the second image data of the sintered particles and calculating a second powder removal rate according to the three-dimensional model includes: Processing the multi-angle images to obtain three-dimensional point cloud data of the sintered particles; Establishing a three-dimensional model of the sintered particles according to the three-dimensional point cloud data, and calculating the volume of the three-dimensional model according to the three-dimensional model; The volume loss rate of the sintered particles is obtained by comparing the volume of the three-dimensional model with a reference model without de-powdering, and the volume loss rate is used as the second de-powdering rate.

6. The method for detecting the drying powder loss rate according to claim 5, wherein: The step of processing the multi-angle images to obtain three-dimensional point cloud data of the sintered particles includes: Preprocessing the multi-angle image to obtain first point cloud data; determining a voxel size according to the point cloud density of the first point cloud data; The first point cloud data is registered based on a normal distribution transformation algorithm and the voxel size to obtain the three-dimensional point cloud data.

7. The method for detecting the drying powder loss rate according to claim 1, wherein: The third image data of the sintered particles is analyzed to obtain a third powder removal rate, including: The motion trajectory of the sintered particles is extracted by the inter-frame difference method; the trajectory curvature and motion trajectory are analyzed and identified to obtain the key frame position where the collision event occurs; According to the key frame position in the time sequence window before and after the collision event, the ROI area of the sintered particles is sub-pixel registered, and the grayscale reduction of the surface pixel grayscale value of the sintered particles before and after the collision is calculated; When the grayscale reduction is greater than the set threshold, it is determined to be a powder-shedding area; and the third powder-shedding rate is calculated based on the ratio of the area of the powder-shedding area to the reference surface area of the sintered particles.

8. A drying powder loss rate detection device, characterized in that: include: a first data processing module, configured to process first image data of the sintered particles to obtain a first powder removal rate, wherein the first image data is an RGB image of the sintered particles; a second data processing module, configured to obtain a three-dimensional model of the sintered particles based on second image data of the sintered particles, and calculate a second powder removal rate based on the three-dimensional model, wherein the second image data is a multi-angle image of the sintered particles; a third data processing module, configured to analyze third image data of the sintered particles to obtain a third powder removal rate, wherein the third image data is a continuous frame image of the sintered particles; The fourth data processing module is used to perform weighted calculation on the first powder removal rate, the second powder removal rate, and the third powder removal rate to obtain a target powder removal rate.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images