A disinfection equipment control method and system based on object recognition
By collecting the image of the item in the ultraviolet disinfection equipment and calculating its volume and surface roughness, and automatically adjusting the disinfection time, the problem of incomplete disinfection or excessive disinfection in the existing technology is solved, the disinfection accuracy and efficiency are improved, and energy waste is avoided Item damage.
Patent Information
- Application Number
- CN202411113789.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-14
AI Technical Summary
The existing ultraviolet disinfection equipment does not consider the size or surface characteristics of the items to be disinfected during disinfection, which will lead to incomplete disinfection or excessive disinfection, affect the disinfection effect, and may cause energy waste and damage to the items.
By collecting images of the items to be disinfected before and after the UV disinfection light is turned on, peeling off the background area, and calculating the volume and surface roughness of the items, so as to automatically adjust the disinfection time of the disinfection equipment to achieve refined disinfection.
It improves disinfection accuracy, avoids unnecessary energy consumption and damage to items, and realizes personalized disinfection according to the situation of the items.
Smart Images

Figure CN118642417B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image processing, and in particular to a disinfection equipment control method and system based on object recognition. Background Art
[0002] As society pays more and more attention to health and hygiene, disinfection equipment is increasingly used in medical, food processing, home and other public places. Ultraviolet (UV) disinfection lamps are a common disinfection tool that is widely used because they can effectively kill bacteria and viruses. However, existing UV disinfection equipment usually uses a fixed disinfection time without considering factors such as the size or surface characteristics of the items to be disinfected. This method may lead to incomplete disinfection or excessive disinfection, which not only affects the disinfection effect, but may also cause energy waste and damage to items. Summary of the invention
[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a disinfection equipment control method and system based on object recognition to solve the problem of low disinfection accuracy of existing ultraviolet disinfection equipment.
[0004] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a disinfection equipment control method based on object recognition, comprising the following steps:
[0005] S1. Place the items to be disinfected in the disinfection area;
[0006] S2. Before the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a bird's-eye view to obtain a first image;
[0007] S3, after the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a top-down angle to obtain a second image;
[0008] S4, stripping the background region of the first image to obtain a first object image, and stripping the background region of the second image to obtain a second object image;
[0009] S5. Obtaining the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image;
[0010] S6. Calculate the disinfection time of the disinfection equipment based on the volume and surface roughness of the items.
[0011] Furthermore, the background stripping of the first image and the background stripping of the second image in S4 both include the following sub-steps:
[0012] S41, calculating the similarity between the pixel value of each pixel point in the original image and the background pixel value, wherein the original image is the first image or the second image;
[0013] S42, when the similarity is greater than the similarity threshold, the corresponding pixel point is marked as a background pixel point;
[0014] S43, removing background pixels from the original image to obtain an object image.
[0015] Furthermore, the S5 comprises the following sub-steps:
[0016] S51, subtracting pixel values of the first object image and the second object image at the same pixel points to obtain a difference image, wherein the pixel value of each position point in the difference image is: , r i is the pixel value of the i-th position point in the difference image, P 1,i is the pixel value of the i-th pixel on the first object image, P 2,i is the pixel value of the i-th pixel on the second object image, where i is a positive integer;
[0017] S52, classifying the difference image to obtain various classification areas on the difference image;
[0018] S53, calculating the height of each classification area on the difference image;
[0019] S54, calculating the volume of the object according to the area of each classification region and the height of each classification region on the difference image;
[0020] S55. Calculate the surface roughness of the object according to each classification area on the difference image.
[0021] Further, the S52 includes the following sub-steps:
[0022] S521, taking any unclassified position point on the difference image as a classification point;
[0023] S522, calculating the pixel value distance between the unclassified position point and the classified point within the neighborhood range of the classified point;
[0024] S523, determine whether there is an unclassified position point in the neighborhood whose corresponding pixel value distance is less than the distance threshold, if so, classify the classified point and the unclassified position point in the neighborhood into one classified area, and jump to step S524, if not, when the classified point has a classified area, directly jump to step S524, when the classified point does not have a classified area, classify the classified point into one classified area, and jump to step S521;
[0025] S524. Select any position point from the classification area that has not been classified as a new classification point, and jump to step S522 until all the position points in the classification area have been classified points. The classification area is classified, and jump to step S521 until the position points on the difference image are classified to obtain each classification area on the difference image.
[0026] Furthermore, the S53 comprises the following sub-steps:
[0027] S531, calculating the average pixel value of each classification area on the difference image;
[0028] S532: Calculate the height of each classification area according to the average pixel value of the classification area.
[0029] Furthermore, the formula for calculating the height of the classification area in S532 is: , where h is the height of the classification area and θ is the proportional coefficient. is the average pixel value of the classification area, ε is the adjustment coefficient;
[0030] The formula for calculating the volume of the item in S54 is: , where V is the volume of the item, E j is the area of the jth classification region on the gap image, h j is the height of the jth classification region on the difference image, j is a positive integer, and M is the number of classification regions.
[0031] Furthermore, the S55 comprises the following sub-steps:
[0032] S551, dividing the difference image into a plurality of regions of the same size;
[0033] S552, calculating the roughness of each area according to the average pixel value of each classification area in each area;
[0034] S553. Select the maximum roughness from the roughness of each area as the surface roughness of the object.
[0035] Furthermore, the formula for calculating the roughness of the area in S552 is: , where γ is the roughness of the area, is the average pixel value of the nth classified area in the region, L is the number of classified areas in the region, n is a positive integer, and e is a natural constant.
[0036] Furthermore, the formula for calculating the disinfection time of the disinfection equipment in S6 is: , where T is the disinfection time of the disinfection equipment, α 1 is the first weighting coefficient, α 2is the second weighting coefficient, V is the volume of the object, γ is the roughness of the area, V th is the volume threshold, γ th is the roughness threshold.
[0037] A disinfection equipment control system based on object recognition, comprising: a placement unit, a first acquisition unit, a second acquisition unit, a background stripping unit, an acquisition unit and a duration calculation unit;
[0038] The placement unit is used to place the items to be sterilized in the sterilization area;
[0039] The first acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle before the ultraviolet disinfection lamp is turned on to obtain a first image;
[0040] The second acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle after the ultraviolet disinfection lamp is turned on, to obtain a second image;
[0041] The background stripping unit is used to strip the background area of the first image to obtain the first object image, and strip the background area of the second image to obtain the second object image;
[0042] The acquisition unit is used to acquire the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image;
[0043] The duration calculation unit is used to calculate the disinfection duration of the disinfection equipment according to the volume of the object and the surface roughness of the object.
[0044] In summary, the beneficial effects of the present invention are as follows: the present invention places the objects to be disinfected in the disinfection area, collects images before and after the ultraviolet disinfection lamp is turned on, and strips off the background area, retaining only the object image, to obtain a first object image and a second object image, and obtains the object volume and the object surface roughness according to the difference between the first object image and the second object image, thereby calculating the disinfection time of the disinfection equipment, realizing automatic refined disinfection according to the situation of the objects, and improving the disinfection accuracy of the objects. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 The present invention is a flow chart of a disinfection equipment control method based on object recognition. DETAILED DESCRIPTION
[0046] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0047] Example 1: Figure 1 As shown, a disinfection equipment control method based on object recognition includes the following steps:
[0048] S1. Place the items to be disinfected in the disinfection area;
[0049] S2. Before the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a bird's-eye view to obtain a first image;
[0050] S3, after the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a top-down angle to obtain a second image;
[0051] S4, stripping the background region of the first image to obtain a first object image, and stripping the background region of the second image to obtain a second object image;
[0052] S5. Obtaining the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image;
[0053] S6. Calculate the disinfection time of the disinfection equipment based on the volume and surface roughness of the items.
[0054] Ultraviolet disinfection lamps usually appear as light blue or blue-purple light on objects. Ultraviolet disinfection lamps, also known as ultraviolet germicidal lamps, mainly emit short-wave ultraviolet rays (UVC) with a wavelength of 253.7 nanometers, which has a strong bactericidal ability. Although ultraviolet rays themselves are invisible light, because low-pressure mercury lamps also produce a small amount of visible light, mainly blue light and purple light, while emitting ultraviolet light, ultraviolet disinfection lamps appear light blue. Therefore, the present invention uses the difference in the image before and after the ultraviolet disinfection lamp is turned on to obtain the volume of the object and the surface roughness of the object.
[0055] In S4, both the background stripping of the first image and the background stripping of the second image include the following sub-steps:
[0056] S41, calculating the similarity between the pixel value of each pixel point in the original image and the background pixel value, wherein the original image is the first image or the second image;
[0057] S42, when the similarity is greater than the similarity threshold, the corresponding pixel point is marked as a background pixel point;
[0058] S43, removing background pixels from the original image to obtain an object image.
[0059] The present invention calculates the similarity between the pixel value of each pixel in the original image and the background pixel value. When the similarity is greater than a similarity threshold, the pixel value is similar to that of the pixel in the background, and the pixel is marked as a background pixel. The background pixel is removed from the original image to obtain an object image.
[0060] The background pixel value is obtained by extracting the pixel value in the background area image before and after the ultraviolet disinfection lamp is turned on.
[0061] The S5 comprises the following sub-steps:
[0062] S51, subtracting pixel values of the first object image and the second object image at the same pixel points to obtain a difference image, wherein the pixel value of each position point in the difference image is: , r i is the pixel value of the i-th position point in the difference image, P 1,i is the pixel value of the i-th pixel on the first object image, P 2,i is the pixel value of the i-th pixel on the second object image, where i is a positive integer;
[0063] S52, classifying the difference image to obtain various classification areas on the difference image;
[0064] S53, calculating the height of each classification area on the difference image;
[0065] S54, calculating the volume of the object according to the area of each classification region and the height of each classification region on the difference image;
[0066] S55. Calculate the surface roughness of the object according to each classification area on the difference image.
[0067] The present invention performs pixel value subtraction on the first object image and the second object image, and the obtained difference image reflects the influence of the ultraviolet disinfection lamp on the disinfected object. The area with large pixel values on the difference image has a large influence of the ultraviolet disinfection lamp on the disinfected object and is adequately irradiated by the ultraviolet disinfection lamp. The area with small pixel values on the difference image has a small influence of the ultraviolet disinfection lamp on the disinfected object. The present invention classifies the difference image according to the pixel value, thereby realizing the classification of the difference image according to the influence, and each classified area reflects the irradiation effect of the ultraviolet disinfection lamp.
[0068] The S52 comprises the following sub-steps:
[0069] S521, taking any unclassified position point on the difference image as a classification point;
[0070] S522, calculating the pixel value distance between the unclassified position point and the classified point within the neighborhood range of the classified point;
[0071] In this embodiment, the pixel value distance may be the absolute value of the difference between the two pixel values;
[0072] S523, determine whether there is an unclassified position point in the neighborhood whose corresponding pixel value distance is less than the distance threshold, if so, classify the classified point and the unclassified position point in the neighborhood into one classified area, and jump to step S524, if not, when the classified point has a classified area, directly jump to step S524, when the classified point does not have a classified area, classify the classified point into one classified area, and jump to step S521;
[0073] Starting from step S521, during the first traversal, if there is no position point with a pixel value less than the distance threshold in the neighborhood of the classification point, the classification point is classified as a classification area. After entering step S524, the classification point selected from the classification area has a corresponding classification area, and there is no position point with a pixel value less than the distance threshold in the neighborhood of the classification point, and the process directly enters step S524 to select the next classification point.
[0074] S524. Select any position point from the classification area that has not been classified as a new classification point, and jump to step S522 until all the position points in the classification area have been classified points. The classification area is classified, and jump to step S521 until the position points on the difference image are classified to obtain each classification area on the difference image.
[0075] When classifying, the present invention does not simply classify the position points belonging to the same pixel value range together, but searches for position points with similar pixel values to the classification points through the neighborhood range of the classification points, thereby classifying them into one category, and then selects position points that have not been classified points from the classification area as new classification points, and searches for position points with similar pixel values to the new classification points from the neighborhood range of the new classification points, until all the position points in the classification area have been classified points, the classification of the classification area is completed, and then a new classification area is searched, so that adjacent position points with similar pixel values are classified into one category, the classification accuracy is improved, and the influence of the ultraviolet disinfection lamp on each area of the difference image is reflected in a refined manner.
[0076] The S53 comprises the following sub-steps:
[0077] S531, calculating the average pixel value of each classification area on the difference image;
[0078] S532: Calculate the height of each classification area according to the average pixel value of the classification area.
[0079] The formula for calculating the height of the classification area in S532 is: , where h is the height of the classification area and θ is the proportional coefficient. is the average pixel value of the classification area, and ε is the adjustment coefficient.
[0080] In this embodiment, the proportional coefficient is used to configure the ratio of the height to the average pixel value, and the adjustment coefficient is used to configure the ratio The remaining value after .
[0081] The closer the object is to the ultraviolet disinfection lamp, the greater the impact of the ultraviolet disinfection lamp on the object. Therefore, the present invention uses the average pixel value of each classification area on the difference image to estimate the height of the classification area.
[0082] The formula for calculating the volume of the item in S54 is: , where V is the volume of the item, E j is the area of the jth classification region on the gap image, h j is the height of the jth classification region on the difference image, j is a positive integer, and M is the number of classification regions.
[0083] The present invention obtains the volume of the object by comprehensively considering the area of each classification region and the height of the classification region.
[0084] The S55 comprises the following sub-steps:
[0085] S551, dividing the difference image into a plurality of regions of the same size;
[0086] S552, calculating the roughness of each area according to the average pixel value of each classification area in each area;
[0087] S553. Select the maximum roughness from the roughness of each area as the surface roughness of the object.
[0088] The formula for calculating the roughness of the area in S552 is: , where γ is the roughness of the area, is the average pixel value of the nth classified area in the region, L is the number of classified areas in the region, n is a positive integer, and e is a natural constant.
[0089] In this embodiment, step S551 specifically includes: cutting the difference image into The roughness of each area is calculated based on the average pixel value of each classified area in each area. The greater the roughness, the more irregular the surface of the object. Therefore, the largest roughness is selected. The largest roughness represents the area of the object that is difficult to disinfect.
[0090] When calculating the roughness of the area, the average pixel value of each classification area and the overall average pixel value are used. The larger the gap is and the more classification areas there are in the area, the greater the roughness of the area.
[0091] The formula for calculating the disinfection time of the disinfection equipment in S6 is: , where T is the disinfection time of the disinfection equipment, α 1 is the first weighting coefficient, α 2 is the second weighting coefficient, V is the volume of the object, γ is the roughness of the area, V th is the volume threshold, γ th is the roughness threshold.
[0092] The present invention sets thresholds for the volume of the object and the roughness of the area respectively. When the volume of the object is greater than the volume threshold and the roughness of the area is greater than the roughness threshold, the disinfection time of the disinfection equipment is longer.
[0093] In this embodiment, the threshold and the threshold are specifically set according to requirements or experiments.
[0094] Embodiment 2: A disinfection equipment control system based on object recognition, comprising: a placement unit, a first acquisition unit, a second acquisition unit, a background stripping unit, an acquisition unit and a duration calculation unit;
[0095] The placement unit is used to place the items to be sterilized in the sterilization area;
[0096] The first acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle before the ultraviolet disinfection lamp is turned on to obtain a first image;
[0097] The second acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle after the ultraviolet disinfection lamp is turned on, to obtain a second image;
[0098] The background stripping unit is used to strip the background area of the first image to obtain the first object image, and strip the background area of the second image to obtain the second object image;
[0099] The acquisition unit is used to acquire the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image;
[0100] The duration calculation unit is used to calculate the disinfection duration of the disinfection equipment according to the volume of the object and the surface roughness of the object.
[0101] The specific implementation process of Example 2 is the same as that of Example 1.
[0102] The present invention places the objects to be disinfected in a disinfection area, collects images before and after the ultraviolet disinfection lamp is turned on, and strips off the background area to retain only the object image, thereby obtaining a first object image and a second object image, and according to the difference between the first object image and the second object image, obtains the object volume and the object surface roughness, thereby calculating the disinfection time of the disinfection equipment, realizing automatic and refined disinfection according to the situation of the objects, and improving the disinfection accuracy of the objects.
[0103] The present invention acquires the volume and surface roughness information of the object by collecting images of the object to be disinfected before and after the ultraviolet disinfection lamp is turned on, and processing the images to remove the background area. This information is then used to calculate the required disinfection time to achieve targeted disinfection. The present invention can not only improve the targeting and efficiency of disinfection, but also avoid unnecessary energy consumption and damage to objects.
[0104] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A disinfection equipment control method based on object recognition, characterized in that: The following steps are involved: S1. Place the items to be disinfected in the disinfection area; S2. Before the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a bird's-eye view to obtain a first image; S3, after the ultraviolet disinfection lamp is turned on, an image of the object to be disinfected is collected from a top-down angle to obtain a second image; S4, stripping the background region of the first image to obtain a first object image, and stripping the background region of the second image to obtain a second object image; S5. Obtaining the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image; S6. Calculate the disinfection time of the disinfection equipment according to the volume of the items and the surface roughness of the items; The S5 comprises the following sub-steps: S51, subtracting pixel values of the first object image and the second object image at the same pixel points to obtain a difference image, wherein the pixel value of each position point in the difference image is: , r i is the pixel value of the i-th position point in the difference image, P 1,i is the pixel value of the i-th pixel on the first object image, P 2,i is the pixel value of the i-th pixel on the second object image, where i is a positive integer; S52, classifying the difference image to obtain various classification areas on the difference image; S53, calculating the height of each classification area on the difference image; S54, calculating the volume of the object according to the area of each classification region and the height of each classification region on the difference image; S55, calculating the surface roughness of the object according to each classification area on the difference image; The S53 comprises the following sub-steps: S531, calculating the average pixel value of each classification area on the difference image; S532, calculating the height of each classification area according to the average pixel value of the classification area; The formula for calculating the height of the classification area in S532 is: , where h is the height of the classification area and θ is the proportional coefficient. is the average pixel value of the classification area, ε is the adjustment coefficient; The formula for calculating the volume of the item in S54 is: , where V is the volume of the item, E j is the area of the jth classification region on the gap image, h j is the height of the jth classification region on the difference image, j is a positive integer, and M is the number of classification regions; The S55 comprises the following sub-steps: S551, dividing the difference image into a plurality of regions of the same size; S552, calculating the roughness of each area according to the average pixel value of each classification area in each area; S553, selecting the maximum roughness from the roughness of each area as the surface roughness of the object; The formula for calculating the roughness of the area in S552 is: , where γ is the roughness of the area, is the average pixel value of the nth classified area in the region, L is the number of classified areas in the region, n is a positive integer, and e is a natural constant.
2. The disinfection equipment control method based on object recognition according to claim 1 is characterized in that: In S4, both the background stripping of the first image and the background stripping of the second image include the following sub-steps: S41, calculating the similarity between the pixel value of each pixel point in the original image and the background pixel value, wherein the original image is the first image or the second image; S42, when the similarity is greater than the similarity threshold, the corresponding pixel point is marked as a background pixel point; S43, removing background pixels from the original image to obtain an object image.
3. The disinfection equipment control method based on object recognition according to claim 1 is characterized in that: The S52 comprises the following sub-steps: S521, taking any unclassified position point on the difference image as a classification point; S522, calculating the pixel value distance between the unclassified position point and the classified point within the neighborhood range of the classified point; S523, determine whether there is an unclassified position point in the neighborhood whose corresponding pixel value distance is less than the distance threshold, if so, classify the classified point and the unclassified position point in the neighborhood into one classified area, and jump to step S524, if not, when the classified point has a classified area, directly jump to step S524, when the classified point does not have a classified area, classify the classified point into one classified area, and jump to step S521; S524. Select any position point from the classification area that has not been classified as a new classification point, and jump to step S522 until all the position points in the classification area have been classified points. The classification area is classified, and jump to step S521 until the position points on the difference image are classified to obtain each classification area on the difference image.
4. The disinfection equipment control method based on object identification according to claim 1 is characterized in that: The formula for calculating the disinfection time of the disinfection equipment in S6 is: , where T is the disinfection time of the disinfection equipment, α1 is the first weighting coefficient, α2 is the second weighting coefficient, V is the volume of the object, γ is the roughness of the area, V th is the volume threshold, γ th is the roughness threshold.
5. A disinfection equipment control system based on object recognition, implemented based on the disinfection equipment control method based on object recognition according to any one of claims 1 to 4, characterized in that: include: A placement unit, a first acquisition unit, a second acquisition unit, a background stripping unit, an acquisition unit and a duration calculation unit; The placement unit is used to place the items to be sterilized in the sterilization area; The first acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle before the ultraviolet disinfection lamp is turned on to obtain a first image; The second acquisition unit is used to acquire an image of the object to be disinfected from a top-down angle after the ultraviolet disinfection lamp is turned on, to obtain a second image; The background stripping unit is used to strip the background area of the first image to obtain the first object image, and strip the background area of the second image to obtain the second object image; The acquisition unit is used to acquire the volume of the object and the surface roughness of the object according to the difference between the first object image and the second object image; The duration calculation unit is used to calculate the disinfection duration of the disinfection equipment according to the volume of the object and the surface roughness of the object.
Citation Information
Patent Citations
Multi-spectrum-based disinfection effect detection method, disinfection control method and equipment
CN114235718A
Luggage disinfection device and control method thereof
CN115147581A
Method, device, and program for measuring number of object, recording medium recorded with the program
JP2010262397A
Movable Sterilizer possible setting the sterilization time using Automatic volume calculation.
KR1020150077713A
Ultraviolet disinfection system
US20170290933A1
