An intelligent system for food testing

Abstract

The present invention relates to the field of detection, and in particular to an intelligent system for food detection. The present invention is provided with a transport detection mechanism including a conveyor belt and a grabbing mechanical arm, a detection module including an X-ray scanning unit, and a food analyzer including a food classification unit and a shaking analysis unit, to obtain a scanned image of a packaged food to be detected during shaking, extract contour features in the scanned image, and then identify specific features, determine the category of the packaged food to be detected in combination with the contour area of ​​the specific features, determine non-specific features in each adjacent scanned image frame, determine whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features, identify fragment features based on the difference in contour features in the scanned image, determine the aggregation area of ​​the fragment features in each scanned image, and determine whether the packaged food meets the standard based on the area of ​​the largest aggregation area, thereby improving the detection reliability and detection accuracy of the packaged food to be detected.

Description

Technical Field

[0001] The present invention relates to the field of food detection, and in particular to an intelligent system applied to food detection. Background Art

[0002] Food testing, as an important means to improve product quality and market competitiveness, plays an important role in ensuring food safety, improving quality of life, promoting market supervision, protecting consumer rights, and promoting industry development. In addition, with the development of science and technology, the application of intelligent rapid testing technologies and platforms is also gradually being promoted. The application of these technologies has provided new means for food safety supervision, improved supervision efficiency and transparency, and helped to promptly discover and deal with food-related issues.

[0003] Chinese patent publication number: CN117358601A, discloses a food finished metal foreign body detection device, including a conveyor belt and a metal detector, etc.; the metal detector is installed in the middle of the conveyor belt. The metal foreign body slides down the arc groove along with the grains by squeezing the screening plate through the squeezing block, and then the separated grains are recycled and reprocessed through the discharge groove of the conveying frame, reducing resource waste, and the grains without metal foreign bodies are prevented from falling into the arc groove by squeezing the baffle plate and the squeezing plate, thereby improving the metal foreign body screening effect.

[0004] In the prior art, the inspection of packaged food is mostly aimed at the outer packaging, ingredients and microbial distribution, and the internal state of the packaged food is not fully inspected. However, in actual inspections, due to falling, squeezing, collisions, etc., the packaged food is likely to be internally damaged, causing the food to break, affecting the internal state of the packaged food and affecting the comprehensiveness, reliability and accuracy of the inspection. Summary of the invention

[0005] To this end, the present invention provides an intelligent system for food detection, which is used to overcome the problem that the prior art does not take into account the possibility that packaged food may be internally damaged due to falling, squeezing, etc., resulting in food breakage, affecting the internal state of the packaged food, and affecting the comprehensiveness, reliability and accuracy of the detection.

[0006] To achieve the above object, the present invention provides an intelligent system for food detection, which comprises:

[0007] A transport inspection mechanism, comprising a transport belt for transporting the packaged food to be inspected and a grabbing mechanical arm arranged on one side of the transport belt for grabbing the packaged food to be inspected;

[0008] A detection module, comprising an X-ray scanning unit disposed on one side of the conveyor belt for obtaining a scanning image of the packaging bag;

[0009] A food analyzer, which is connected to the transport detection mechanism and the detection module respectively, comprises a food classification unit and a shaking analysis unit,

[0010] The food classification unit is used to control the grasping mechanical arm to grasp the packaged food to be detected and move back and forth to make the packaged food to be detected shake, obtain a scanned image of the packaged food to be detected during the shaking process, extract contour features in the scanned image, identify specific features based on the displacement change of the contour features in adjacent scanned image frames, and determine the category of the packaged food to be detected based on the contour area of ​​the specific features;

[0011] The shaking analysis unit is used to analyze each scanned image based on the determination result of the food classification unit to determine whether the packaged food meets the standard, including:

[0012] Determine non-specific features in each adjacent scanned image frame, and determine whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features;

[0013] Alternatively, fragment features are identified based on differences in contour features in the scanned images, clustered areas of the fragment features in each scanned image are determined, and whether the packaged food meets the standards is determined based on the area of ​​the largest clustered area.

[0014] Furthermore, the food classification unit is used to identify the specific features based on the displacement change of the contour features in the adjacent scanned image frames, including:

[0015] for identifying contour features in each scanned image frame corresponding to adjacent scanned images;

[0016] Used to determine the coordinate change distance of the midpoint of a single contour feature in adjacent scanned image frames;

[0017] Determining the coordinate change distance as the displacement change amount of the contour feature;

[0018] It is used to identify the contour features whose displacement change is less than a predetermined displacement change threshold as special features.

[0019] Furthermore, the food classification unit determines the category of the packaged food to be detected based on the contour area of ​​the specific feature, wherein:

[0020] If the contour area of ​​the unique feature is greater than or equal to the preset reference contour area, the category of the packaged food to be tested is determined to be non-dispersed packaged food;

[0021] If the contour area of ​​the unique feature is smaller than the preset reference contour area, the category of the packaged food to be inspected is determined to be dispersed packaged food.

[0022] Furthermore, the shaking analysis unit determines whether the packaged food meets the standard based on the determination result of the food classification unit, wherein:

[0023] If the category of packaged food is non-dispersed packaged food, calculate the ratio of the contour area of ​​non-specific features to specific features to determine whether the packaged food meets the standards;

[0024] If the packaged food is classified as dispersed packaged food, the area of ​​the largest aggregation region is calculated to determine whether the packaged food meets the standards.

[0025] Furthermore, the jitter analysis unit calculates the contour area ratio of the non-specific feature to the specific feature based on each adjacent scan image, including:

[0026] To obtain adjacent scan images;

[0027] To calculate the average area of ​​specific feature contours and the average area of ​​non-specific feature contours;

[0028] The ratio of the average area of ​​the specific feature to the average area of ​​the non-specific feature contour is determined as the contour area ratio.

[0029] Furthermore, the shaking analysis unit determines whether the packaged food meets the standard based on the contour area ratio of the non-specific feature to the specific feature.

[0030] If the ratio of the contour area of ​​the non-specific features to the specific features is greater than the reference contour area ratio, the packaged food is judged to be non-compliant;

[0031] If the ratio of the contour areas of non-specific features to specific features is less than or equal to the reference contour area ratio, the packaged food is judged to meet the standards.

[0032] Furthermore, the jitter analysis unit identifies the fragment features based on the difference of the contour features in the scanned image, including:

[0033] to determine the area ratio of the area of ​​each contour feature in the scanned image to the average area;

[0034] Comparing the area ratio corresponding to each contour feature with a preset reference area ratio, and determining the fragment feature according to the comparison result;

[0035] If the area ratio of a contour feature is smaller than a preset reference area ratio, the contour is identified as a fragment feature.

[0036] Furthermore, the jitter analysis unit determines the clustering area of ​​the fragment features in each scanned image, including:

[0037] A feature set for identifying aggregated fragments in a scanned image;

[0038] Determine the area where the clustered fragment feature set is located as the clustering area;

[0039] Among them, the clustered fragment feature set is a combination of several fragment features, the number of fragment features in each clustered fragment feature set must be greater than a predetermined number threshold, and the distance between any fragment feature in the clustered fragment feature set and the nearest fragment feature must be less than a predetermined distance threshold.

[0040] Furthermore, the shaking analysis unit determines whether the packaged food meets the standard according to the area of ​​the largest aggregation area, including:

[0041] To compare the areas of the aggregation areas of the fragment features in each scanned image to determine the maximum aggregation area;

[0042] to compare the area of ​​the maximum aggregation region with a preset aggregation area ratio, and determine whether the packaged food meets the standards based on the comparison result;

[0043] Among them, if the area of ​​the largest aggregation region is greater than the preset aggregation area ratio, it is determined that the packaged food does not meet the standards.

[0044] Furthermore, it also includes a display module for displaying the image detected by the detection module.

[0045] Compared with the prior art, the present invention sets up a transport detection mechanism including a conveyor belt and a grabbing robot arm, a detection module including an X-ray scanning unit, and a food analyzer including a food classification unit and a shaking analysis unit. The present invention obtains a scanned image of the packaged food to be detected during shaking, extracts contour features in the scanned image, and then identifies specific features, determines the category of the packaged food to be detected in combination with the contour area of ​​the specific features, determines non-specific features in each adjacent scanned image frame, determines whether the packaged food meets the standard based on the ratio of the contour area of ​​the non-specific features to the specific features, identifies fragment features based on the difference in contour features in the scanned image, determines the aggregation area of ​​the fragment features in each scanned image, and determines whether the packaged food meets the standard according to the area of ​​the largest aggregation area, thereby improving the detection reliability and accuracy of the packaged food to be detected.

[0046] In particular, the present invention obtains a scanned image of the packaged food to be inspected during the shaking process and identifies specific features. In actual situations, product packaging is diverse. For example, potato chips, rice crusts, etc. are bulk-packed in several pieces, while instant noodles, cakes, and other product packaging only contain a single food item. Different packaging methods have an impact on the analysis method of the scanned image. Therefore, the present invention considers identifying specific features based on the displacement change of contour features in adjacent scanned image frames. For food packaging that only contains a single food item or a small number of larger food items, the main food items are not easy to move when they are grabbed and shaken. Therefore, by identifying the specific features and combining the area of ​​the specific features to identify the above phenomenon, a basis is provided for the subsequent classification of the packaged food to be inspected, and different image analysis methods are adaptively adopted to thereby improve the detection reliability and accuracy of the packaged food to be inspected.

[0047] In particular, the present invention adopts a dynamic analysis method for non-dispersed packaged foods. Since this type of packaged food to be tested only contains a single food piece or a small number of larger food pieces, the main food pieces are not easy to move. If food fragments are present, the movement amplitude of the food fragments relative to the main food pieces will be relatively large. This type of dynamic features is easier to identify, the data representation is strong, and the dynamic movement process will be relatively prominent in multiple scanned images. Therefore, the corresponding non-specific features are used as the basis to determine whether the packaged food meets the standards, thereby improving the detection reliability and accuracy of the packaged food to be tested.

[0048] In particular, for dispersed packaged foods, an overall analysis method is adopted, which is subject to the influence of the accuracy of the scanned image. Under normal circumstances, the food pieces in the dispersed packages are mixed with broken food fragments, and it is not easy to highlight the unclear boundaries in the scanned image. Therefore, the present invention considers shaking the packaged food to be inspected and collecting several scanned images during the shaking process. Due to the influence of the shaking, the broken food fragments are easy to deposit on one side of the packaging bag. The deposited features are more prominent in the scanned image. Therefore, the present invention considers determining the aggregation area of ​​the fragment features to determine whether the packaged food to be inspected meets the standards, thereby improving the detection reliability and accuracy of the packaged food to be inspected. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the structure of an intelligent system for food detection according to an embodiment of the invention;

[0050] Figure 2 A logic block diagram for determining the category of packaged food to be tested according to an embodiment of the invention;

[0051] Figure 3 A logic block diagram for determining whether packaged food meets the standards according to an embodiment of the invention;

[0052] Figure 4 This is a logic block diagram of an embodiment of the invention for determining whether dispersed packaged food meets standards. DETAILED DESCRIPTION

[0053] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0054] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0055] It should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] See also Figure 1-Figure 4 As shown, Figure 1 is a schematic diagram of the structure of an intelligent system applied to food detection according to an embodiment of the present invention, Figure 2 is a logic block diagram of determining the category of packaged food to be detected according to an embodiment of the present invention. Figure 3 This is a logic block diagram of determining whether non-dispersed packaged food meets the standards according to an embodiment of the present invention. Figure 4 This is a logic block diagram of an embodiment of the invention for determining whether dispersed packaged food meets the standards. The intelligent system for food detection of the present invention includes:

[0057] A transport inspection mechanism, comprising a transport belt for transporting the packaged food to be inspected and a grabbing mechanical arm arranged on one side of the transport belt for grabbing the packaged food to be inspected;

[0058] A detection module, comprising an X-ray scanning unit disposed on one side of the conveyor belt for obtaining a scanning image of the packaging bag;

[0059] A food analyzer, which is connected to the transport detection mechanism and the detection module respectively, comprises a food classification unit and a shaking analysis unit,

[0060] The food classification unit is used to control the grasping mechanical arm to grasp the packaged food to be detected and move back and forth to make the packaged food to be detected shake, obtain a scanned image of the packaged food to be detected during the shaking process, extract contour features in the scanned image, identify specific features based on the displacement change of the contour features in adjacent scanned image frames, and determine the category of the packaged food to be detected in combination with the contour area of ​​the specific features;

[0061] The shaking analysis unit is used to analyze each scanned image based on the determination result of the food classification unit to determine whether the packaged food meets the standard, including:

[0062] Determine non-specific features in each adjacent scanned image frame, and determine whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features;

[0063] Alternatively, fragment features are identified based on differences in contour features in the scanned images, clustered areas of the fragment features in each scanned image are determined, and whether the packaged food meets the standards is determined based on the area of ​​the largest clustered area.

[0064] Specifically, there is no limitation on the specific material of the conveyor belt, which can be rubber, fiber, metal, plastic, etc. Technical personnel in this field can select suitable conveyor belt materials according to different usage environments and needs. Preferably, cold-resistant conveyor belts can be selected in cold areas, and heat-resistant conveyor belts can be selected in hot areas. This will not be repeated.

[0065] Specifically, there is no limitation on the specific type of the grasping robot arm, which can be a linear arm, a SCARA arm, a joint multi-axis robot arm, etc. The degree of freedom of the robot arm only needs to meet the usage requirements and it only needs to be able to grasp and shake the bagged food. This is the existing technology and will not be elaborated.

[0066] Specifically, there is no limitation on the specific form of the X-ray scanning unit, which can be an X-ray scanning device that uses X-ray digital radiography inspection technology. It only needs to be able to scan packaged food to obtain a scanned image. This is a prior art, and those skilled in the art can choose it according to specific needs and conditions, which will not be elaborated here.

[0067] Specifically, there is no limitation on the shaking amplitude and frequency of the packaged food, and those skilled in the art can make their own judgments in advance based on the image scanning results. It is sufficient to ensure that the specific features can be scanned relatively clearly and that subsequent analysis can be performed, and this will not be elaborated here.

[0068] Specifically, there is no limitation on the specific structure of the food analyzer and each unit therein, and they can all be composed of logic components or a combination of logic components. The logic components include a field programmable processor, a computer, or a microprocessor in a computer.

[0069] Specifically, the food classification unit is used to identify specific features based on the displacement change of contour features in adjacent scanned image frames, including:

[0070] for identifying contour features in each scanned image frame corresponding to adjacent scanned images;

[0071] It is used to determine the coordinate change distance of the midpoint of a single contour feature in adjacent scanned image frames. The coordinate change distance is the relative distance between two coordinates and can be calculated according to the coordinate distance formula;

[0072] Determining the coordinate change distance as the displacement change amount of the contour feature;

[0073] It is used to identify the contour features whose displacement change is less than a predetermined displacement change threshold as special features.

[0074] Adjacent scanned images are the basic standard for calculation. In some possible implementations, it can be increased to a group of four frames. The displacement change after the increase should be the average of the displacement changes corresponding to the three adjacent scanned images in the four consecutive scanned images. This will improve the accuracy of the displacement change, but will reduce the computing power of the computer. Those skilled in the art will make adjustments based on actual conditions, which will not be repeated here.

[0075] Specifically, there is no limitation on the method of identifying contour features, and image segmentation algorithms may be used. It is understandable that the corresponding algorithms or image processing models may be imported into the logic components in advance to process the images and realize the corresponding functions, which will not be elaborated here.

[0076] Specifically, there is no limitation on the method for determining the midpoint of a single contour feature. It can be determined by coordinate positioning or by using an image processing algorithm. This is a prior art and will not be described in detail.

[0077] Specifically, the predetermined displacement change threshold D0 is obtained by pre-statistical analysis, wherein a number of scanned images are obtained, and the average displacement change ΔD of the contour feature in each adjacent scanned image is obtained, and D0=ΔD×g is set, where g represents a correction coefficient, 0.85<g<0.95.

[0078] Specifically, the food classification unit determines the category of the packaged food to be detected based on the contour area of ​​the specific feature, wherein:

[0079] If the contour area of ​​the unique feature is greater than or equal to the preset reference contour area, the category of the packaged food to be tested is determined to be non-dispersed packaged food;

[0080] If the contour area of ​​the unique feature is smaller than the preset reference contour area, the category of the packaged food to be inspected is determined to be dispersed packaged food.

[0081] Specifically, the preset reference contour area is 0.6 to 0.8 times the area of ​​the packaged food bag to be inspected, and those skilled in the art can set it according to their needs.

[0082] The present invention obtains a scanned image of the packaged food to be detected during the shaking process and identifies specific features. In actual situations, product packaging is diverse. For example, potato chips, rice crusts, etc. are bulk-packed in several pieces, while instant noodles, cakes and other product packaging only contain a single food item. Different packaging methods have an impact on the analysis method of the scanned image. Therefore, the present invention considers identifying specific features by based on the displacement change of contour features in adjacent scanned image frames. For food packaging containing only a single food item or a small number of larger food items, the main food items are not easy to move when they are grabbed and shaken. Therefore, by identifying the specific features and combining the area of ​​the specific features to identify the above phenomenon, a basis is provided for the subsequent classification of the packaged food to be detected, and different image analysis methods are adaptively adopted, thereby improving the detection reliability and accuracy of the packaged food to be detected.

[0083] Specifically, the shaking analysis unit determines whether the packaged food meets the standard based on the determination result of the food classification unit, wherein:

[0084] If the category of packaged food is non-dispersed packaged food, calculate the ratio of the contour area of ​​non-specific features to specific features to determine whether the packaged food meets the standards;

[0085] If the packaged food is classified as dispersed packaged food, the area of ​​the largest aggregation region is calculated to determine whether the packaged food meets the standards.

[0086] Specifically, the jitter analysis unit calculates the contour area ratio of the non-specific feature to the specific feature based on each adjacent scan image, including:

[0087] To obtain adjacent scan images;

[0088] To calculate the average area of ​​specific feature contours and the average area of ​​non-specific feature contours;

[0089] The ratio of the average contour area of ​​the non-specific features to the average contour area of ​​the specific features is determined as the contour area ratio.

[0090] It is understandable that each adjacent scan image is at least two groups to calculate the average value to ensure the accuracy of detection. Those skilled in the art can determine the specific number of each adjacent scan image according to actual conditions, which will not be repeated here.

[0091] Specifically, the shaking analysis unit determines whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features.

[0092] If the ratio of the contour area of ​​the non-specific features to the specific features is greater than the reference contour area ratio, the packaged food is judged to be non-compliant;

[0093] If the ratio of the contour areas of non-specific features to specific features is less than or equal to the reference contour area ratio, the packaged food is judged to meet the standards.

[0094] Specifically, the reference profile area ratio is determined within the interval [0.05, 0.2], and those skilled in the art may select it according to actual conditions, which will not be elaborated here.

[0095] It is understandable that the ratio of the contour area of ​​the non-specific feature to the specific feature should be less than 1. Specifically, if the ratio is greater than 1, in order to ensure the accuracy of the detection, this step of calculation should be performed again.

[0096] The present invention adopts a dynamic analysis method for non-dispersed packaged foods. Since this type of packaged food to be tested only contains a single food item or a small number of larger food items, the main food items are not easy to move. If there are food fragments, the movement amplitude of the food fragments relative to the main food items will be relatively large. This type of dynamic features are easier to identify and have strong data representation. The dynamic movement process will be relatively prominent in multiple scanned images. Therefore, the corresponding non-specific features are used as the basis to determine whether the packaged food meets the standards, thereby improving the detection reliability and accuracy of the packaged food to be tested.

[0097] Specifically, the shaking analysis unit identifies the fragment features based on the difference of contour features in the scanned image, including:

[0098] to determine the area ratio of the area of ​​each contour feature in the scanned image to the average area;

[0099] Comparing the area ratio corresponding to each contour feature with a preset reference area ratio, and determining the fragment feature according to the comparison result;

[0100] If the area ratio of a contour feature is smaller than a preset reference area ratio, the contour is identified as a fragment feature.

[0101] Specifically, the preset reference area ratio is selected within the interval [0.03, 0.16], and those skilled in the art may select it according to actual conditions, which will not be elaborated here.

[0102] Specifically, the jitter analysis unit determines the clustering area of ​​the fragment features in each scanned image, including:

[0103] A feature set for identifying aggregated fragments in a scanned image;

[0104] Determine the area where the clustered fragment feature set is located as the clustering area;

[0105] Among them, the clustered fragment feature set is a combination of several fragment features, the number of fragment features in each clustered fragment feature set must be greater than a predetermined number threshold, and the distance between any fragment feature in the clustered fragment feature set and the nearest fragment feature must be less than a predetermined distance threshold.

[0106] Specifically, the predetermined quantity threshold is 3, and the predetermined distance threshold is selected within the interval [0.01A, 0.03A], where A is the length of the short side of the packaged food bag. Those skilled in the art may select it according to actual conditions, which will not be elaborated here.

[0107] Specifically, the shaking analysis unit determines whether the packaged food meets the standard according to the area of ​​the largest aggregation area, including:

[0108] To compare the areas of the aggregation areas of the fragment features in each scanned image to determine the maximum aggregation area;

[0109] to compare the area of ​​the maximum aggregation region with a preset aggregation area ratio, and determine whether the packaged food meets the standards based on the comparison result;

[0110] Among them, if the area of ​​the largest aggregation region is greater than the preset aggregation area ratio, it is determined that the packaged food does not meet the standards.

[0111] It is understandable that shaking the packaged food multiple times will produce multiple scanned images, and the clustered areas scanned each time may be inconsistent. The fragment features are aggregated for easy identification, so the area of ​​the largest clustered area corresponding to the scanned image is considered as the judgment basis.

[0112] For dispersed packaged foods, an overall analysis method is adopted, which is subject to the influence of the accuracy of the scanned image. Under normal circumstances, the food pieces in the dispersed packages are mixed with broken food fragments, and it is not easy to highlight the unclear boundaries in the scanned image. Therefore, the present invention considers shaking the packaged food to be inspected and collecting several scanned images during the shaking process. Due to the influence of the shaking, the broken food fragments are easy to deposit on one side of the packaging bag. The deposited features are more prominent in the scanned image. Therefore, the present invention considers determining the aggregation area of ​​the fragment features to determine whether the packaged food to be inspected meets the standards, thereby improving the detection reliability and accuracy of the packaged food to be inspected.

[0113] Specifically, it also includes a display module for displaying the image detected by the detection module.

[0114] Specifically, there is no limitation on the specific form of the display module, and those skilled in the art can flexibly adjust it according to specific application scenarios and user needs. Preferably, in order to make the image clear, a high-resolution display can be selected, which will not be elaborated here.

[0115] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An intelligent system for food detection, characterized in that: include: A transport inspection mechanism, comprising a transport belt for transporting the packaged food to be inspected and a grabbing mechanical arm arranged on one side of the transport belt for grabbing the packaged food to be inspected; A detection module, comprising an X-ray scanning unit disposed on one side of the conveyor belt for obtaining a scanning image of the packaging bag; A food analyzer, which is connected to the transport detection mechanism and the detection module respectively, comprises a food classification unit and a shaking analysis unit, The food classification unit is used to control the grasping mechanical arm to grasp the packaged food to be detected and move back and forth to make the packaged food to be detected shake, obtain a scanned image of the packaged food to be detected during the shaking process, extract contour features in the scanned image, identify specific features based on the displacement change of the contour features in adjacent scanned image frames, and determine the category of the packaged food to be detected based on the contour area of ​​the specific features; The shaking analysis unit is used to analyze each scanned image based on the determination result of the food classification unit to determine whether the packaged food meets the standard, including: Determine non-specific features in each adjacent scanned image frame, and determine whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features; or, identifying fragment features based on differences in contour features in the scanned images, determining the clustering area of ​​the fragment features in each scanned image, and determining whether the packaged food meets the standards based on the area of ​​the largest clustering area; The food classification unit is used to identify specific features based on the displacement change of contour features in adjacent scanned image frames, including: for identifying contour features in each scanned image frame corresponding to adjacent scanned images; Used to determine the coordinate change distance of the midpoint of a single contour feature in adjacent scanned image frames; Determining the coordinate change distance as the displacement change amount of the contour feature; It is used to identify the contour features whose displacement change is less than a predetermined displacement change threshold as special features.

2. The intelligent system for food detection according to claim 1, characterized in that: The food classification unit determines the category of the packaged food to be detected based on the contour area of ​​the specific feature, wherein: If the contour area of ​​the unique feature is greater than or equal to the preset reference contour area, the category of the packaged food to be tested is determined to be non-dispersed packaged food; If the contour area of ​​the unique feature is smaller than the preset reference contour area, the category of the packaged food to be inspected is determined to be dispersed packaged food.

3. The intelligent system for food detection according to claim 1, characterized in that: The shaking analysis unit determines whether the packaged food meets the standard based on the determination result of the food classification unit, wherein: If the category of packaged food is non-dispersed packaged food, calculate the ratio of the contour area of ​​non-specific features to specific features to determine whether the packaged food meets the standards; If the packaged food is classified as dispersed packaged food, the area of ​​the largest aggregation region is calculated to determine whether the packaged food meets the standards.

4. The intelligent system for food detection according to claim 1, characterized in that: The shaking analysis unit calculates the contour area ratio of the non-specific feature to the specific feature based on each adjacent scan image, including: To obtain adjacent scan images; To calculate the average area of ​​specific feature contours and the average area of ​​non-specific feature contours; The ratio of the average contour area of ​​the non-specific features to the average contour area of ​​the specific features is determined as the contour area ratio.

5. The intelligent system for food detection according to claim 1, characterized in that: The shaking analysis unit determines whether the packaged food meets the standard based on the contour area ratio of the non-specific features to the specific features. If the ratio of the contour area of ​​the non-specific features to the specific features is greater than the reference contour area ratio, the packaged food is judged to be non-compliant; If the ratio of the contour areas of non-specific features to specific features is less than or equal to the reference contour area ratio, the packaged food is judged to meet the standards.

6. The intelligent system for food detection according to claim 1, characterized in that: The shaking analysis unit identifies the debris features based on the difference of the contour features in the scanned image, including: to determine the area ratio of the area of ​​each contour feature in the scanned image to the average area; Comparing the area ratio corresponding to each contour feature with a preset reference area ratio, and determining the fragment feature according to the comparison result; If the area ratio of a contour feature is smaller than a preset reference area ratio, the contour is identified as a fragment feature.

7. The intelligent system for food detection according to claim 6, characterized in that: The jitter analysis unit determines the clustering area of ​​the debris features in each scanned image, including: A feature set for identifying aggregated fragments in a scanned image; Determine the area where the clustered fragment feature set is located as the clustering area; Among them, the clustered fragment feature set is a combination of several fragment features, the number of fragment features in each clustered fragment feature set must be greater than a predetermined number threshold, and the distance between any fragment feature in the clustered fragment feature set and the nearest fragment feature must be less than a predetermined distance threshold.

8. The intelligent system for food detection according to claim 7, characterized in that: The shaking analysis unit determines whether the packaged food meets the standard according to the area of ​​the largest aggregation area, including: To compare the areas of the aggregation areas of the fragment features in each scanned image to determine the maximum aggregation area; to compare the area of ​​the maximum aggregation region with a preset aggregation area ratio, and determine whether the packaged food meets the standards based on the comparison result; Among them, if the area of ​​the largest aggregation region is greater than the preset aggregation area ratio, it is determined that the packaged food does not meet the standards.

9. The intelligent system for food detection according to claim 1, characterized in that: It also includes a display module for displaying the image detected by the detection module.

Images