A method for detecting the sealing property of a hot pot seasoning packaging bag
Patent Information
- Application Number
- CN202611071917.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-20
- Publication Date
- 2026-08-21
AI Technical Summary
这种方法效率低、主观性强，极易因视觉疲劳导致漏检，尤其对于微小泄漏或不明显的外观形变难以发现
本发明提供的火锅底料包装袋密封性检测方法，通过“色差初筛—激光复检”的两级递进式质检架构，实现了速度与精度的平衡。首先，利用第一运输带上方的摄像头获取包装袋彩色图像，通过构建红油特征区与辣椒特征区，并基于放大后的辣椒特征区修正红油特征区以排除辣椒干扰，再分别计算红油色差评估指数与辣椒色差评估指数，以两者的和值作为疑似漏气的判定依据。这一基于颜色分布统计的初筛步骤能够快速、连续地排除绝大多数密封正常的成品，仅将颜色异常的可疑品导向第二运输带。随后，利用激光扫描头仅对疑似漏气状态的成品进行表面形貌检测，获取以RGB值表征高度信息的第一光谱图样，并通过与质检图样中辣椒特征区的旋转拟合，精准提取辣椒光谱特征区，进而依据该区域的高度分布情况判断表面纹理是保持紧缩状（尖锐折痕和深陷凹坑）还是出现膜浮状（折痕变浅变平），最终确认是否存在封装缺陷。
Smart Images
Figure CN122605741A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of machine vision analysis technology, and more specifically, to a method for detecting the airtightness of hot pot base packaging bags. Background Technology
[0002] As a compound seasoning, the airtightness of hot pot base packaging directly affects the product's shelf life and food safety. Vacuum-packed hot pot base with sealing defects is prone to oil oxidation, mold growth, or oil leakage. Therefore, the sealing quality needs to be checked on the production line.
[0003] Current methods for testing airtightness mainly involve manual sampling and visual inspection, relying on quality inspectors at the end of the production line or after packaging to visually check for obvious leaks in the packaging bags. This method is inefficient, highly subjective, and prone to missed leaks due to visual fatigue, especially for minor leaks or subtle deformations.
[0004] Therefore, there is an urgent need for a method to test the sealing of hot pot base packaging bags that can both ensure detection speed and guarantee that the false negative rate and false positive rate meet production requirements. Summary of the Invention
[0005] The purpose of this invention is to provide a method for testing the sealing performance of hot pot base packaging bags, so as to solve the above-mentioned technical problems.
[0006] To achieve the above objectives, the embodiments of this application provide the following technical solutions: This application provides a method for detecting the airtightness of hot pot base packaging bags, applicable to a transport device with a quality inspection mechanism. The transport device includes a first transport belt and a second transport belt connected to a branch area of the first transport belt. The branch area of the first transport belt is provided with a guiding mechanism for guiding finished hot pot base products determined to be abnormal on the first transport belt to the second transport belt. A camera is located directly above the first transport belt, and a laser scanning head is located on the second transport belt. The method includes: responding to a quality inspection shooting command triggered by an infrared switch located on the first transport belt, the camera located above the first transport belt performs quality inspection shooting. The system obtains the quality inspection image of the hot pot base material currently passing through the first conveyor belt inspection area; it uses a color-based image feature matching algorithm to detect whether there is a color difference in the hot pot base material quality inspection image. If there is a color difference exceeding a first threshold, the current hot pot base material is marked as a suspected leaking state, and a guide delay timer is triggered so that the current hot pot base material reaches the branching area of the first conveyor belt when the delay timer ends. Then, it is transferred to the second conveyor belt through a guide mechanism and a laser scanning head is used for sealing inspection. If the sealing inspection fails, it is determined that the current hot pot base material packaging bag has a sealing defect and is transferred to the recycling box.
[0007] Optionally, a color-based image feature matching algorithm is used to detect whether there is a color difference in the quality inspection sample of hot pot base, including: A color-based image feature matching algorithm is used to construct red oil feature area and chili feature area in the quality inspection image of hot pot base. Then, the chili feature area is magnified to 125%-140% to obtain the magnified chili feature area. The red oil feature area is corrected based on the magnified chili feature area, and multiple first color difference detection points are generated in the corrected red oil feature area based on the square grid sampling algorithm. The first RGB values corresponding to multiple first color difference detection points are calibrated, and the first mean and first standard deviation corresponding to multiple first RGB values are calculated; The first color difference evaluation index is obtained by weighting the difference between the first mean and the preset standard red oil RGB value and the first standard deviation. The first color difference evaluation index is used to characterize the deviation between the red oil color in the corrected red oil feature area and the preset standard red oil color. Calculate the RGB mean value corresponding to the chili pepper feature area, and obtain the second color difference evaluation index based on the difference between the RGB mean value and the preset standard chili pepper RGB value through a weighted algorithm; Calculate the sum of the first color difference assessment index and the second color difference assessment index. If the sum is greater than the first color difference threshold and less than the second color difference threshold, the current hot pot base product is determined to be in a suspected air leakage state, that is, there is a color difference, but it is not significant. If it is greater than or equal to the second color difference threshold, it is directly determined to be a significant color difference.
[0008] Optionally, a sealing quality inspection can be performed using a laser scanning head, including: The hot pot base product suspected of being leaking air was scanned by a laser scanning head on the second conveyor belt to obtain the corresponding first spectral pattern. The RGB in the first spectral pattern is used to characterize the height value of the upper surface of the hot pot base product. The quality inspection pattern of hot pot base corresponding to the hot pot base product suspected of leaking air is retrieved, and the location information of the chili characteristic area is marked. Then, based on the location information of the chili characteristic area, rotation fitting is performed on the first spectral pattern, and the chili spectral characteristic area is marked in the first spectral pattern. Based on the spectral distribution of the chili pepper spectral feature region, the surface texture features of the chili pepper packaging area are determined. If the surface texture features are tight, with sharp creases and deep pits, the chili pepper packaging is determined to be normal, and the state of the hot pot base product is changed from a suspected air leakage state to a normal state. If the surface texture features are a floating film, and the creases become shallower or flatter, then the chili pepper packaging is determined to be abnormal, and the status of the hot pot base product is changed from a suspected air leakage state to a packaging defect state.
[0009] The beneficial effects of this invention are as follows: The sealing test method for hot pot base packaging bags provided by this invention achieves a balance between speed and accuracy through a two-stage progressive quality inspection architecture of "color difference initial screening - laser re-inspection". First, a camera above the first conveyor belt acquires a color image of the packaging bag. By constructing a red oil feature area and a chili pepper feature area, and correcting the red oil feature area based on the magnified chili pepper feature area to eliminate chili pepper interference, the red oil color difference evaluation index and the chili pepper color difference evaluation index are calculated separately. The sum of the two is used as the basis for judging suspected leakage. This initial screening step based on color distribution statistics can quickly and continuously eliminate most of the finished products with normal sealing, and only guide the suspected products with abnormal colors to the second conveyor belt. Subsequently, a laser scanning head is used to inspect the surface morphology of the finished products suspected of leakage, and obtain a first spectral pattern with RGB values representing height information. By rotating and fitting the chili pepper feature area in the quality inspection pattern, the chili pepper spectral feature area is accurately extracted. Then, based on the height distribution of this area, it is determined whether the surface texture remains compressed (sharp creases and deep pits) or has a floating film (creases become shallower and flatter), and finally confirms whether there is a packaging defect.
[0010] By magnifying the chili pepper feature area and correcting the red oil feature area, interference from chili pepper fragments on the color extraction of the red oil area is effectively avoided, making color difference assessment more accurate. Laser scanning re-inspection directly analyzes the height texture of the chili pepper area, judging whether the packaging film adheres tightly to the material from the perspective of physical deformation. It can reliably distinguish between sealed qualified products and defective products that float on the surface due to micro-leakage, completely eliminating the hidden dangers of false sealing and micro-leakage that cannot be identified by relying solely on color.
[0011] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description
[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 This is a schematic diagram of a method for testing the sealing performance of hot pot base packaging bags as described in an embodiment of the present invention. Detailed Implementation
[0014] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
[0015] Example 1
[0016] like Figure 1 As shown, this embodiment provides a method for detecting the sealing performance of hot pot base packaging bags, applicable to a transport device with a quality inspection mechanism. The transport device includes a first transport belt and a second transport belt connected to a bifurcation area of the first transport belt. The bifurcation area of the first transport belt is equipped with a guiding mechanism for guiding the hot pot base products determined to be abnormal on the first transport belt to the second transport belt. A camera is located directly above the first transport belt, equipped with a ring light source to ensure uniform illumination. A laser scanning head is located on the second transport belt, capable of acquiring the relative height information of each point on the surface of the material and outputting it in the form of a pseudo-color image (i.e., a first spectral pattern), wherein different RGB pixel values correspond to different height values. The method is characterized by comprising: Step S100: When the hot pot base product moves to the infrared switch position along the first conveyor belt, the hot pot base product blocks the infrared light, and the infrared switch immediately generates an electrical pulse signal. This signal is sent to the camera controller as a quality inspection shooting command. In response to the quality inspection shooting command triggered by the infrared switch on the first conveyor belt, the camera above the first conveyor belt performs quality inspection shooting to obtain the quality inspection image of the hot pot base product passing through the quality inspection area of the first conveyor belt. For better appearance, two decorative chili peppers are placed on the surface of each hot pot base product. In the event of slight air leakage, the chili pepper feature area and the red oil feature area will change color to a certain extent. However, there is an error in the color change detection based on the image. Therefore, for hot pot base products with insignificant color change, it is necessary to further detect the sealing and tightening texture of the chili pepper feature area (a more accurate judgment criterion). For hot pot base products with significant color change, the guide mechanism will directly put them into the recycling box.
[0017] Step S200: Based on the color-based image feature matching algorithm, detect whether there is a color difference in the quality inspection image of the hot pot base. If there is a color difference exceeding the first threshold, mark the current hot pot base as a suspected air leak and trigger the guide delay timer so that the current hot pot base reaches the first conveyor belt branching area when the delay timer ends. Then, it is transferred to the second conveyor belt through the guide mechanism and the sealing quality is checked by the laser scanning head. If the sealing quality check fails, it is determined that the current hot pot base packaging bag has a sealing defect and is transferred to the recycling box.
[0018] The implementation method of detecting color difference in the hot pot base finished product quality inspection sample using the color-based image feature matching algorithm in step S200 can be as follows: Step S210: Based on the color class image feature matching algorithm, construct the red oil feature area and the chili feature area in the hot pot base finished product quality inspection image, and then enlarge the chili feature area to 125%-140% to obtain the enlarged chili feature area; Step S220: Correct the red oil feature area based on the magnified chili feature area, and generate multiple first color difference detection points in the corrected red oil feature area based on the square grid sampling algorithm; Step S230: Calibrate the first RGB values corresponding to multiple first color difference detection points, and calculate the first mean and first standard deviation corresponding to multiple first RGB values; Step S240: Based on the difference between the first mean and the preset standard red oil RGB value and the first standard deviation, a first color difference evaluation index is obtained through a weighted algorithm. The first color difference evaluation index is used to characterize the deviation between the red oil color in the corrected red oil feature area and the preset standard red oil color. Step S250: Calculate the RGB mean value corresponding to the chili pepper feature area, and obtain the second color difference evaluation index based on the difference between the RGB mean value and the preset standard chili pepper RGB value through a weighted algorithm; Step S260: Calculate the sum of the first color difference evaluation index and the second color difference evaluation index, and determine that the current hot pot base product is in a suspected air leakage state when the sum is greater than the first color difference threshold and less than the second color difference threshold.
[0019] Secondly, the specific implementation method of performing sealing quality inspection by laser scanning head in step S200 can be as follows: Step S270: Scan the hot pot base product that is suspected of leaking air on the second conveyor belt using a laser scanning head to obtain the corresponding first spectral pattern. The RGB values in the first spectral pattern are used to characterize the height value of the upper surface of the hot pot base product. Step S280: Retrieve the hot pot base finished product quality inspection pattern corresponding to the hot pot base product suspected of leaking air, and mark the location information of the chili characteristic area. Then, based on the location information of the chili characteristic area, perform rotation fitting on the first spectral pattern, and then mark the chili spectral characteristic area in the first spectral pattern. Step S290: Based on the spectral distribution of the chili pepper spectral feature region, determine the surface texture features of the chili pepper packaging area. If the surface texture features are tight, with sharp creases and deep pits, then the chili pepper packaging is normal, and the state of the hot pot base product is changed from a suspected air leakage state to a normal state. If the surface texture features are a floating film, and the creases become shallower or flatter, then the chili pepper packaging is determined to be abnormal, and the status of the hot pot base product is changed from a suspected air leakage state to a packaging defect state.
[0020] The hot pot base packaging bag sealing detection method provided in this embodiment achieves a balance between speed and accuracy through a two-stage progressive quality inspection architecture of "color difference initial screening - laser re-inspection". First, a camera above the first conveyor belt acquires a color image of the packaging bag. By constructing a red oil feature area and a chili pepper feature area, and correcting the red oil feature area based on the magnified chili pepper feature area to eliminate chili pepper interference, the red oil color difference evaluation index and the chili pepper color difference evaluation index are calculated separately. The sum of the two is used as the basis for judging suspected leakage. This initial screening step based on color distribution statistics can quickly and continuously eliminate most of the finished products with normal sealing, and only guide the suspicious products with abnormal colors to the second conveyor belt. Subsequently, a laser scanning head is used to detect the surface morphology of the finished products suspected of leakage, and obtain a first spectral pattern with RGB values representing height information. By rotating and fitting the chili pepper feature area in the quality inspection pattern, the chili pepper spectral feature area is accurately extracted. Then, based on the height distribution of this area, it is determined whether the surface texture remains tight (sharp creases and deep pits) or has a floating film (creases become shallower and flatter), and finally confirms whether there is a packaging defect.
[0021] By magnifying the chili pepper feature area and correcting the red oil feature area, interference from chili pepper fragments on the color extraction of the red oil area is effectively avoided, making color difference assessment more accurate. Laser scanning re-inspection directly analyzes the height texture of the chili pepper area, judging whether the packaging film adheres tightly to the material from the perspective of physical deformation. It can reliably distinguish between sealed qualified products and defective products that float on the surface due to micro-leakage, completely eliminating the hidden dangers of false sealing and micro-leakage that cannot be identified by relying solely on color.
[0022] Example 2
[0023] This embodiment, based on Embodiment 1, further illustrates the specific implementation method for determining the surface texture features of the chili pepper packaging area: The RGB values of each pixel within the spectral feature region of the chili pepper are extracted and converted into height values to generate a height map of the chili pepper region. Calculate the corresponding root mean square height based on the height map of the chili pepper region; Calculate the height change rate of each pixel along the X and Y directions within a preset unit pixel step. and Then, the gradient magnitude is synthesized. ; In a typical vacuum-packed bag, the film adheres tightly to the hard chili pepper, and the "wrinkles" on its surface appear as sharply undulating "ridges" and "valleys" on a height map. A common characteristic of these locations is their gradient amplitude. It will be significantly larger than the surrounding flat area and form a narrow "ridge" or "valley" outline.
[0024] Key pixel selection method: Traverse the height map of the chili pepper region; if the gradient magnitude of a pixel is... If both of the following conditions are met, then it is marked as a key pixel: Condition 1: The gradient magnitude of the pixel is greater than the preset gradient magnitude threshold. The preset gradient magnitude threshold is based on the statistics of normal qualified products and represents the minimum slope that distinguishes between "flat film surface" and "wrinkled edge".
[0025] Condition 2: Within a certain neighborhood perpendicular to the gradient direction, the pixel... It must be the maximum value. This ensures that only points on the "ridge" or "valley" of the fold are taken, rather than the gentle slope area at the edge of the fold.
[0026] The points selected in this way are precisely the locations where the wrinkles and textures are most prominent and the curvature features are most concentrated. For a normal, compact chili pepper area, the number of key points selected accounts for about 5% to 15% of the total pixels. This can directly exclude more than the majority of pixels in the relatively flat areas, preventing them from participating in subsequent complex curvature calculations.
[0027] For each of the key pixels selected in the previous step, perform curvature calculations: For each key pixel, take out an m×m small window centered on it, and then calculate the first and second partial derivatives of that point using the difference approximation formula; The average absolute value of curvature is synthesized based on the first-order and second-order partial derivatives, and this average absolute value of curvature is used as a measure of the "sharpness" of the key pixel. The percentage of key pixels with an average absolute curvature greater than a preset curvature threshold is counted and recorded as the key pixel density. This value reflects the proportion of truly "sharp" folds on the fold outline itself. After leakage, the folds become rounded and blunt, and this proportion decreases significantly. In the case of slight leakage, the proportion of "sharp" folds decreases significantly, while the key pixels, i.e., the fold outline itself, do not change much. If the key pixels, i.e., the fold outline itself, also decrease significantly, this is considered a more significant leakage, which can usually be detected and identified as a leakage state in the early color difference detection.
[0028] If the density of key pixels is lower than a preset density threshold, it is determined that there is air leakage in the chili pepper area. The preset density threshold is also obtained through statistical analysis of multiple normal samples.
[0029] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for detecting the sealing performance of hot pot base packaging bags, applicable to a transport device equipped with a quality inspection mechanism, the transport device comprising a first transport belt and a second transport belt connected to a bifurcation area of the first transport belt, the bifurcation area of the first transport belt being provided with a guiding mechanism for guiding finished hot pot base products determined to be abnormal on the first transport belt to the second transport belt, a camera being provided directly above the first transport belt, and a laser scanning head being provided on the second transport belt, characterized in that... The method includes: In response to the quality inspection shooting command triggered by the infrared switch located on the first conveyor belt, the camera located above the first conveyor belt performs quality inspection shooting to obtain the quality inspection pattern of the hot pot base material passing through the quality inspection area of the first conveyor belt at that time. An image feature matching algorithm based on color is used to detect whether there is a color difference in the quality inspection image of the hot pot base. If there is a color difference exceeding the first threshold, the current hot pot base is marked as a suspected air leak and a guide delay timer is triggered so that the current hot pot base reaches the branching area of the first conveyor belt when the delay timer ends. Then, it is transferred to the second conveyor belt through the guide mechanism and a laser scanning head is used for sealing inspection. If the sealing inspection fails, it is determined that the current hot pot base packaging bag has a sealing defect and is transferred to the recycling box.
2. The method for testing the sealing performance of hot pot base packaging bags according to claim 1, characterized in that, Image feature matching algorithms based on color class are used to detect color differences in finished hot pot base samples, including: A color-based image feature matching algorithm is used to construct red oil feature area and chili feature area in the quality inspection image of hot pot base. Then, the chili feature area is magnified to 125%-140% to obtain the magnified chili feature area. The red oil feature area is corrected based on the magnified chili feature area, and multiple first color difference detection points are generated in the corrected red oil feature area based on the square grid sampling algorithm. The first RGB values corresponding to multiple first color difference detection points are calibrated, and the first mean and first standard deviation corresponding to multiple first RGB values are calculated; The first color difference evaluation index is obtained by weighting the difference between the first mean and the preset standard red oil RGB value and the first standard deviation. The first color difference evaluation index is used to characterize the deviation between the red oil color in the corrected red oil feature area and the preset standard red oil color. Calculate the RGB mean value corresponding to the chili pepper feature area, and obtain the second color difference evaluation index based on the difference between the RGB mean value and the preset standard chili pepper RGB value through a weighted algorithm; Calculate the sum of the first color difference assessment index and the second color difference assessment index, and determine that the current hot pot base product is in a suspected air leakage state when the sum is greater than the first color difference threshold and less than the second color difference threshold.
3. The method for testing the sealing performance of hot pot base packaging bags according to claim 2, characterized in that, Sealing quality inspection is performed using a laser scanning head, including: The hot pot base product suspected of being leaking air was scanned by a laser scanning head on the second conveyor belt to obtain the corresponding first spectral pattern. The RGB in the first spectral pattern is used to characterize the height value of the upper surface of the hot pot base product. The quality inspection pattern of hot pot base corresponding to the hot pot base product suspected of leaking air is retrieved, and the location information of the chili characteristic area is marked. Then, based on the location information of the chili characteristic area, rotation fitting is performed on the first spectral pattern, and the chili spectral characteristic area is marked in the first spectral pattern. Based on the spectral distribution of the chili pepper spectral feature region, the surface texture features of the chili pepper packaging area are determined. If the surface texture features are tight, with sharp creases and deep pits, the chili pepper packaging is determined to be normal, and the state of the hot pot base product is changed from a suspected air leakage state to a normal state. If the surface texture features are a floating film, and the creases become shallower or flatter, then the chili pepper packaging is determined to be abnormal, and the status of the hot pot base product is changed from a suspected air leakage state to a packaging defect state.