5S management system for chemical enterprises

Abstract

The embodiment of the invention provides a 5S management system for a chemical enterprise. The 5S management system for the chemical enterprise comprises a photographing indicator arranged in an image acquisition area, and the photographing indicator is at least used for indicating a photographing position and a photographing requirement; the image acquisition equipment is configured to acquire a target image which is shot from the shooting position, aims at the image acquisition area and meets the shooting requirement; the computing device is configured to determine the matching degree between the target image and a pre-stored reference image of the image acquisition area, and the reference image is an image when the image acquisition area meets the 5S management standard; and performing conversion in a corresponding mode according to the matching degree to obtain a score of the image acquisition area under the 5S management standard.

Description

Technical Field

[0001] The embodiments of this specification generally relate to the technical field of automated implementation of chemical enterprise management, and specifically to a 5S management system for chemical enterprises. Background Art

[0002] In chemical enterprises, good management is essential for production safety, product quality and enterprise efficiency. 5S management (sort, straighten, sweep, clean, and maintain) has been widely used in various enterprises as an effective management method. However, the 5S management of traditional chemical enterprises mainly relies on manual inspection and evaluation, which has problems such as strong subjectivity, low efficiency, and difficulty in comprehensive coverage. At the same time, manual inspection may lead to inaccurate evaluation results due to inconsistent experience and standards of inspectors. With the development of the Internet of Things and computer technology, a more objective, efficient and accurate 5S management system is needed to meet the management needs of chemical enterprises. Summary of the invention

[0003] In view of the above, the embodiment of this specification provides a 5S management system for chemical enterprises. The 5S management system includes: a photographing indicator set in the image acquisition area, wherein the photographing indicator is at least used to indicate the shooting position and shooting requirements; an image acquisition device, configured to obtain a target image taken from the shooting position, for the image acquisition area, and meeting the shooting requirements; and a computing device, configured to determine the degree of match between the target image and a pre-stored reference image of the image acquisition area, wherein the reference image is an image of the image acquisition area when it meets the 5S management standard; according to the size of the matching degree, a corresponding method is used for conversion to obtain a score of the image acquisition area under the 5S management standard. The present invention not only reduces the subjectivity of manual inspection, makes the scoring results more objective, and greatly improves the evaluation efficiency of 5S management, but also carefully distinguishes the differences in 5S management levels under different matching degrees, which helps to achieve refined management. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] A further understanding of the nature and advantages of the content of this specification may be achieved by referring to the following drawings. In the drawings, similar components or features may have the same reference numerals.

[0005] Figure 1 An exemplary architecture of a 5S management system for a chemical enterprise according to an embodiment of the present specification is shown.

[0006] Figure 2 A schematic diagram showing an example of a 5S management system for a chemical enterprise according to an embodiment of the present specification.

[0007] Figure 3A flowchart showing an example of a process for determining a matching degree according to an embodiment of the present specification.

[0008] Figure 4 A flowchart showing another example of a process for determining a matching degree according to an embodiment of the present specification.

[0009] Figure 5 A flowchart showing an example of a process of determining a score of an image acquisition area under the 5S management standard according to an embodiment of the present specification.

[0010] Figure 6 A signaling diagram showing an example of an interaction process of a 5S management system according to an embodiment of this specification. DETAILED DESCRIPTION

[0011] The subject matter described herein will be discussed below with reference to example implementations. It should be understood that the discussion of these implementations is only to enable those skilled in the art to better understand and implement the subject matter described herein, and is not a limitation of the scope of protection, applicability or examples set forth in the claims. The functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the embodiments of this specification. Various processes or components may be omitted, substituted or added to each example as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0012] As used herein, the term "including" and its variations represent open terms, meaning "including but not limited to". The term "based on" means "based at least in part on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other definitions may be included below, whether explicit or implicit. Unless the context clearly indicates otherwise, the definition of a term is consistent throughout the specification.

[0013] The flowcharts used in this specification illustrate the operations implemented by the system according to some embodiments in this specification. It should be clearly understood that the operations of the flowcharts may not be implemented in sequence. On the contrary, the operations may be implemented in reverse order or simultaneously. In addition, one or more other operations may be added to the flowchart. One or more operations may be removed from the flowchart.

[0014] In view of this, the embodiment of this specification proposes a 5S management system for chemical enterprises. The 5S management system for chemical enterprises includes: a photographing indicator set in the image acquisition area, wherein the photographing indicator is at least used to indicate the shooting position and shooting requirements; an image acquisition device, configured to obtain a target image taken from the shooting position, for the image acquisition area, and meeting the shooting requirements; and a computing device, configured to determine the degree of match between the target image and a pre-stored reference image of the image acquisition area, wherein the reference image is an image of the image acquisition area when it meets the 5S management standard; according to the size of the matching degree, a corresponding method is used for conversion to obtain the score of the image acquisition area under the 5S management standard. The 5S management system according to the present invention not only reduces the subjectivity of manual inspection, makes the scoring results more objective, and greatly improves the evaluation efficiency of 5S management, but also carefully distinguishes the differences in 5S management levels under different matching degrees, which is helpful to achieve refined management.

[0015] The 5S management system for a chemical enterprise according to an embodiment of the present specification will be described in detail below with reference to the accompanying drawings.

[0016] Figure 1 An exemplary architecture 100 of a 5S management system for a chemical enterprise according to an embodiment of the present specification is shown.

[0017] exist Figure 1 In the embodiment, the photographing indicator 110 provided in the image acquisition area can at least be used to indicate the shooting position and the shooting requirements. The image acquisition device 120 can be configured to obtain a target image that meets the shooting requirements and is taken from the shooting position and is for the image acquisition area. For example, the image acquisition device 120 can be a camera, a smart phone, a patrol robot with an image acquisition function, etc. The image acquisition device 120 and the computing device 130 can be interconnected through a network. The network can be any type of network that can interconnect network entities, such as a single network or a combination of various networks. In terms of coverage, the network can be a local area network (LAN), a wide area network (WAN), etc. In terms of the carrier medium, the network can be a wired network, a wireless network, etc. In terms of data exchange technology, the network can be a circuit switching network, a packet switching network, etc.

[0018] The computing device 130 may be configured to determine the degree of match between the target image and a pre-stored reference image of the image acquisition area, and then convert the target image in a corresponding manner according to the degree of match to obtain a score of the image acquisition area under the 5S management standard. The reference image may be an image of the image acquisition area when it meets the 5S management standard.

[0019] In one embodiment, the image acquisition device 120 may be used by a user. In some cases, the user may use the image acquisition device 120 to capture a target image according to the requirements of the photographing indicator. Afterwards, the image acquisition device 120 may interact with the computing device 130. For example, the image acquisition device 120 may transmit the target image to the computing device 130. The computing device 130 may generate an associated response based on the received target image, such as a score of the image acquisition area under the 5S management standard.

[0020] It should be understood that Figure 1 All network entities shown in the figure are exemplary. Any other network entities may be involved in the architecture 100 according to specific application requirements.

[0021] Figure 2 A schematic diagram showing an example of a 5S management system 200 for a chemical enterprise according to an embodiment of the present specification.

[0022] like Figure 2 As shown, the 5S management system 200 for a chemical enterprise may include a photographing indicator 210 disposed in an image acquisition area, an image acquisition device 220 and a computing device 230 .

[0023] In this embodiment, the photographing indicator 210 can at least be used to indicate the photographing position and the photographing requirements. In some examples, the photographing indicator can be a signboard with obvious markings, and the best shooting position and specific shooting angle requirements can be marked on the signboard, so that the image acquisition device 220 can accurately obtain the target image that meets the shooting requirements. For example, in the warehouse area of ​​a chemical enterprise, a large signboard can be set up, with an arrow indicating that the shooting position should be in the center of the warehouse, and the shooting angle is marked as 60 degrees to the ground, so that the warehouse picture taken can fully display the placement of the goods and the overall layout of the warehouse. In other examples, the photographing indicator can be a ground mark, such as a shooting area range drawn with colored lines and a shooting direction indicated by numbers and arrows, so that the operator of the image acquisition device can quickly and accurately find the correct shooting position according to these ground marks and shoot as required.

[0024] In some examples, the shooting requirements may include shooting parameters used when shooting the reference image. In some examples, the shooting parameters may include but are not limited to at least one of the following: the height of the lens from the ground, the angle between the lens and the ground, and the selected shooting mode (such as automatic mode, macro mode, night mode, etc.). In some examples, the shooting parameters may include but are not limited to at least one of the following: focal length, aperture, exposure time.

[0025] In some examples, the photographing indicator may include an area identification code set in each image acquisition area. For example, the area identification code may be a specific QR code pasted on the corresponding image acquisition area. In one example, the corresponding area identification code may be posted on the wall at the entrance of the material storage area of ​​a chemical enterprise. The area identification code may be parsed by a corresponding QR code recognition device, for example, the name, number, person in charge, and other information of the material storage area may be obtained.

[0026] In some examples, the photographing indicator may include a Radio Frequency Identification (RFID) tag set at a designated shooting position in each image acquisition area. In some examples, the above-mentioned RFID tag may adopt a passive RFID tag, and each RFID tag has unique identification information. For example, the RFID tag can be pre-buried at a specific position on the ground or installed on a corresponding fixed bracket. In one example, in the production area of ​​a chemical enterprise, grooves can be made in the ground and RFID tags can be embedded according to pre-planned shooting positions. Each RFID tag stores detailed location information of the area and shooting parameters such as shooting angle and height that match it. When the corresponding device that supports radio frequency signal interaction is close to the above-mentioned RFID tag, the information stored in the corresponding RFID tag can be obtained.

[0027] In this embodiment, the image acquisition device 220 may be configured to obtain a target image captured from a shooting position and targeting an image acquisition area that meets the shooting requirements. In some examples, an operator may use the image acquisition device 220 to capture the image acquisition area at the shooting position according to the shooting requirements to obtain the target image.

[0028] In some examples, the image acquisition device 220 may be of the same model as the device for shooting the reference image. The image acquisition device 220 may be configured to use shooting parameters and shoot the image acquisition area from the shooting position to obtain the target image. For example, the image acquisition device 220 may be a high-definition industrial camera of the same brand and model as the device for shooting the reference image. For another example, the image acquisition device 220 may be a smart phone of the same brand and model uniformly issued by the chemical enterprise. Thus, under the premise of ensuring that the key hardware parameters of the image acquisition device 220, such as the sensor specifications, pixel size, lens interface, etc., are consistent, the same shooting position and the same shooting requirements are adopted, and the consistency of the captured target image and the reference image in terms of image quality, viewing angle, imaging effect, etc. can be guaranteed to the greatest extent, which is convenient for subsequent accurate comparative analysis.

[0029] In some examples, the image acquisition device 220 may be further configured to scan a region identification code set in an image acquisition region for the target image, and identify the region identification information corresponding to the target image. In some examples, the image acquisition device 220 may be a device capable of parsing the corresponding region identification code, so that the region identification information encoded in the region identification code can be identified.

[0030] In some examples, the image acquisition device may support a Near Field Communication (NFC) function. For example, the image acquisition device 220 may have an RFID tag identifier. When the RFID tag identifier of the image acquisition device 220 gradually approaches the RFID tag and enters the read-write identification range, the scanned RFID tag can be identified, and the regional identification information of the area stored in the RFID tag and the corresponding shooting parameters can be parsed. In an example, in the laboratory area of ​​a chemical enterprise, when the image acquisition device 220 is close to the RFID tag set at the shooting point, the image acquisition device 220 can communicate with the RFID tag through the NFC function, so that the image acquisition device 220 can identify the regional identification information similar to "Laboratory No. 3 Experimental Bench Area" stored in the RFID tag and shooting parameters similar to "shooting height is 1.2 meters, shooting angle is 30 degrees with the ground, automatic shooting mode". Afterwards, the image acquisition device 220 can automatically adjust its own shooting posture and shooting parameters according to these parameters. If the current shooting parameters are adjusted to be consistent with the identified shooting parameters, for example, the shooting posture of the device itself indicates "shooting height is 1.2 meters, shooting angle is 30 degrees with the ground" and the current shooting parameters indicate that it is in "automatic shooting mode", the image acquisition device 220 can shoot the image acquisition area to obtain the target image. If the current shooting parameters are adjusted to be consistent with the identified shooting parameters, for example, the shooting posture of the device itself indicates "shooting height is 1.2 meters, shooting angle is 30 degrees with the ground" and the current shooting parameters indicate that it is in "automatic shooting mode", the image acquisition device 220 can shoot the image acquisition area to obtain the target image. If the current shooting parameters are inconsistent with the identified shooting parameters or the RFID tag is not scanned, the image acquisition device 220 prohibits shooting, for example, does not respond to the click operation of the camera shutter button. This ensures that the image acquisition device can obtain the target image that meets the requirements.

[0031] In this embodiment, the computing device 230 may be configured to perform the following steps S201 - S202 .

[0032] In S201, the matching degree between the target image and a pre-stored reference image of the image acquisition area is determined.

[0033] In this embodiment, the reference image may be an image of the image acquisition area when it complies with the 5S management standard. In some examples, the reference image may correspond to the image acquisition area one by one, that is, different image acquisition areas may correspond to different reference images. For example, the reference image of the production area is different from the reference image of the office area. In some examples, the reference image may have an area identification. For example, for the production site of TPU (Thermoplastic Urethane, thermoplastic polyurethane elastomer), its reference image may present the following scene: the corresponding equipment, tools, and materials are in a reasonable quantity and in a specified position, the cleanliness of the work area and the on-site environment are well controlled, and there are no obvious safety hazards. For another example, for the storage area, its reference image may present the following scene: various materials are in specified positions, the cleanliness is good, and the labels are clear. For another example, for the production workshop, the reference image can show the standard state of neatly arranged equipment, no accumulation of debris, clean ground, and clear identification.

[0034] In this embodiment, reference images of various image acquisition areas may be pre-stored. After acquiring the target image, the reference image corresponding to the target image may be determined according to the image acquisition area targeted by the target image. In some examples, the image acquisition area may be determined by the shooting location information of the target image. In one example, the shooting location information may be specified by the shooting personnel, for example, "multi-function hall on the fourth floor of the office building", "XX extruder", "XX bin", "XX granulator", "XX mixer", etc. may be selected from a drop-down list or pre-entered. In some examples, the shooting location information may be the geotagging of the photo automatically generated by the image acquisition device, that is, the GPS coordinates are embedded in the EXIF ​​data of the photo.

[0035] In this embodiment, the degree of match between the target image and the corresponding reference image determined can be determined in various ways. In some examples, the target image and the corresponding reference image can be preprocessed first, such as grayscale processing, normalization processing, etc. After that, the structural similarity index (SSIM) algorithm of the image can be used to determine the degree of match between the target image and the corresponding reference image. By calculating the similarity between the target image and the reference image in terms of image brightness, contrast, and structural information, a matching value between 0 and 1 is obtained. The closer the value is to 1, the higher the matching degree is, which means that the 5S management situation presented by the target image is closer to the standard state. For example, in the public channel area of ​​a chemical enterprise, if the value obtained by the SSIM algorithm after calculating the target image and the corresponding reference image is 0.85, it means that the overall 5S management situation in the area is good and is more consistent with the standard situation. In some examples, the preprocessed target image can be compared pixel by pixel with the corresponding reference image, and the matching degree between the two is calculated based on the comparison results of each pixel. For example, pixels with similar colors may be given a higher similarity score; pixels with large color differences may be given a lower similarity score. Finally, the similarity scores of all pixels are combined to obtain the overall matching degree between the target image and the corresponding reference image.

[0036] In S202, a corresponding conversion is performed according to the matching degree to obtain a score of the image acquisition area under the 5S management standard.

[0037] In this embodiment, different sizes of matching degrees may correspond to different conversion methods. In some examples, the conversion method may be determined based on the scoring of historical images. In some examples, a large amount of historical data may be collected first, and each piece of historical data may consist of a matching degree determined for a historical image and a corresponding reference image and the corresponding manual scoring data. For example, images of different areas (such as office workstations, production sites, storage locations, laboratory areas, etc.) collected regularly every month in the past year may be selected. For example, assuming that the matching degree value range determined by various methods as described above is between 0.26 and 0.95, the matching degree may be divided into different intervals. For example, 0 to 0.4 is a low matching degree interval, 0.4 to 0.7 is a medium matching degree interval, and 0.7 to 1 is a high matching degree interval. Accordingly, for each matching degree in each interval, there may be a corresponding manual scoring range. For each interval, a regression analysis method may be used, for example, the matching degree may be set as x, the manual scoring may be set as y, and the regression equation y = ax + b may be fitted by the least squares method, that is, the values ​​of the coefficients a and b may be determined. It can be understood that for each interval, there can be a corresponding set of a and b values. After the matching degree is determined in step S201, a corresponding regression equation can be matched according to the interval, so that the matching degree can be converted into a score under the corresponding 5S management standard according to the matched regression equation.

[0038] Continue to refer Figure 3 , Figure 3 FIG. 1 is a flowchart showing an example of a matching degree determination process 300 according to an embodiment of the present specification. The process 300 may be Figure 2 An exemplary implementation of step S201 in .

[0039] like Figure 3 As shown, at 310, feature description information of a reference image is obtained.

[0040] In this embodiment, the feature description information can be used to describe the image features of the reference image. In some examples, the feature description information of the reference image of each image acquisition area can be pre-stored. In some examples, the stored feature description information can be indexed by the area identifier of the corresponding image acquisition area. Thus, the feature description information of the reference image corresponding to the target image can be obtained.

[0041] In some examples, the feature description information may include a feature histogram of the reference image. In some examples, the feature histogram may include, but is not limited to, at least one of the following: a grayscale histogram, a color feature histogram, a gradient histogram (HOG-Histogram of Oriented Gradients), and a local binary pattern (LBP-Local Binary Pattern) histogram.

[0042] In some examples, the feature description information may include multiple feature point descriptors of the reference image. Among them, the feature point descriptors can be used to describe the local features of the feature points of the reference image. For example, algorithms such as scale-invariant feature transform (SIFT), ORB (Oriented FAST and rotatedBRIEF), and SURF (Speeded up robust feature) can be used to extract feature point descriptors, so that the texture, shape and other features of the local area around each feature point can be obtained. Therefore, a combination of multiple similar feature point descriptors can describe the entire reference image from the perspective of each local feature. It can be understood that when different algorithms are used to obtain the corresponding feature point descriptors for the same feature point, features of different dimensions of the feature point can also be obtained.

[0043] At 320 , feature description information of the target image is extracted according to the method of generating feature description information of the reference image.

[0044] In this embodiment, the feature description information of the target image can be extracted in the same manner as the feature description information of the reference image. In some examples, if the feature description information is a color feature histogram, the same color space division method and interval division rule can be used to count the number of pixels in different color intervals of the target image, thereby generating a color feature histogram of the target image. In some examples, if the feature description information is a texture feature histogram such as a gradient histogram or a local binary pattern histogram, the same texture feature extraction method can be used to perform texture feature analysis on the target image, thereby generating a texture feature histogram of the target image. In some examples, if the feature description information is a plurality of feature point descriptors extracted by the SIFT algorithm, the same SIFT algorithm can be applied, keeping the same parameter settings (such as the number of layers of the Gaussian difference pyramid, the contrast threshold and other parameters unchanged), thereby extracting the feature point descriptors of the target image. In this way, the feature description information of the target image and the reference image is ensured to be comparable and consistent, thereby facilitating the subsequent accurate determination of the degree of match between the two.

[0045] At 330 , a degree of matching between the feature description information of the reference image and the feature description information of the target image is determined.

[0046] In this embodiment, a corresponding matching degree calculation method may be selected according to a specific type of feature description information, so as to determine the matching degree between the feature description information of the reference image and the feature description information of the target image.

[0047] In some examples, if the feature description information is a color feature histogram, the Chi-Square distance, Bhattacharyya distance, histogram intersection area ratio, etc. can be used to measure the similarity between the color feature histogram of the reference image and the color feature histogram of the target image. It can be understood that the smaller the above distance or the larger the histogram intersection area ratio, the more similar the two color feature histograms are, that is, the higher the matching degree. For example, for the image of the raw material stacking area of ​​a chemical enterprise, the Bhattacharyya distance of the color feature histogram of the reference image and the target image can be calculated respectively. If the distance value is small, it means that the target image is closer to the reference image that meets the 5S management standard in terms of color distribution, indicating that the visual presentation of the items in the raw material stacking area, the clean environment, etc. meets the requirements, and the corresponding matching degree is higher. In some examples, if the feature description information is a texture feature histogram, the KL divergence (Kullback-Leibler Divergence) can be used to measure the similarity between the texture feature histogram of the reference image and the texture feature histogram of the target image. It can be understood that the smaller the KL divergence is, the more similar the two texture feature histograms are, that is, the higher the matching degree is.

[0048] Through the above method, this solution can carefully compare the target image and the reference image from the feature level to more accurately determine the difference between the two, thereby providing a reliable basis for the subsequent accurate evaluation of the 5S management situation.

[0049] Reference below Figure 4 , Figure 4 FIG. 4 is a flowchart showing another example of a matching degree determination process 400 according to an embodiment of the present specification. The process 400 may be Figure 3 The process 400 is applicable to the case where the feature description information includes multiple feature point descriptors.

[0050] like Figure 4 As shown, at 410, the similarity between a first number of feature point descriptors of the reference image and a second number of feature point descriptors of the target image is determined.

[0051] In some examples, the feature point descriptor may be a feature vector (e.g., a 128-dimensional vector obtained by an embedding layer). The similarity measure between each feature point descriptor of the reference image and each feature point descriptor of the target image may be determined, thereby obtaining M×N similarity measure values. Among them, M and N may represent a first number and a second number, respectively. In some examples, the similarity measure may take the form of, for example, Euclidean distance or cosine similarity. It can be understood that the smaller the Euclidean distance or the closer the cosine similarity is to 1, the higher the similarity between the two feature point descriptors, that is, the local features of the two corresponding feature points in the image are more consistent.

[0052] At 420 , the number of matched feature point pairs among the first number of feature points in the reference image and the second number of feature points in the target image is determined according to the determined similarity.

[0053] In some examples, it is possible to determine whether two feature points can constitute a matching feature point pair based on a preset similarity threshold. In some examples, if the similarity between a feature point descriptor in the reference image and a feature point descriptor in the target image is greater than a preset threshold (for example, the Euclidean distance is less than a preset Euclidean distance threshold or the cosine similarity is greater than a preset cosine similarity threshold), it can be determined that the feature points indicated by the two feature point descriptors can constitute a matching feature point pair. Similarly, all feature point descriptors in the reference image and the target image can be traversed to count the number of matching feature point pairs. For example, the reference image has 200 feature point descriptors and the target image has 180 feature point descriptors. After the aforementioned similarity judgment, there are 120 pairs of feature point descriptors whose similarity is higher than the set threshold, then it can be determined that the number of matching feature point pairs is 120.

[0054] In some examples, if the similarity between the feature point descriptor of feature point A in the target image and the feature point descriptors of more than one feature point (for example, feature point X' and feature point Y') in the reference image is greater than a preset threshold, only the pair of feature points with the greatest similarity can be determined as a matching feature point pair, that is, each feature point in the target image can constitute at most one feature point pair.

[0055] At 430 , a degree of matching between the feature description information of the reference image and the feature description information of the target image is determined based on at least one of the first number and the second number and the number of matched feature point pairs.

[0056] In this embodiment, the degree of match between the feature description information of the reference image and the feature description information of the target image can be determined in a variety of ways. In some examples, the ratio obtained by dividing the number of matched feature point pairs by a specified value can be determined as the above-mentioned degree of match. The above-mentioned specified value can be determined based on at least one of the first number and the second number. For example, the above-mentioned specified value can be the first number, so that the obtained degree of match is more focused on reflecting the degree of match of the feature descriptor relative to the reference image. For another example, the above-mentioned specified value can be the larger value of the first number and the second number. When the scene in the image acquisition area is more complex or chaotic, for example, when there are more objects (such as objects that do not exist in the reference image), more feature descriptors are often extracted. At this time, the obtained degree of match can reflect the neatness of the current scene to a certain extent.

[0057] In the above manner, this solution can determine the matching degree between the target image and the reference image based on the matching between feature point descriptors, providing a more accurate and rapid image analysis and evaluation method for 5S management of chemical enterprises.

[0058] Reference below Figure 5 , Figure 5 FIG. 5 is a flowchart showing an example of a process 500 for determining a score of an image acquisition area under the 5S management standard according to an embodiment of the present specification. The process 500 may be Figure 3 An exemplary implementation of step S202 in .

[0059] like Figure 5 As shown, at 510, it is determined whether the matching degree is greater than a first preset threshold.

[0060] In this embodiment, the first preset threshold value can be determined according to the value range of the matching degree and the judgment standard for whether the production site has a good site condition under the 5S management standard. In some examples, the value of the first preset threshold value is usually adapted to the value range of the matching degree (e.g., 0~1, or 0~100). In some examples, the judgment standard for whether the production site has a good site condition under the 5S management standard can be determined according to the score for the historical production site image, for example, the last one in the top 30% of the score of the historical production site image can be used as the score line to determine the first preset threshold value. For example, there are 1,000 historical production site images, and the first preset threshold value can be determined according to the score of the 300th historical production site image arranged from high to low according to the score. For example, if the score is 79 and the value range of the matching degree is 0~1, the first preset threshold value can be determined as 7.9.

[0061] If the answer at step 510 is yes, then at step 520 , it is determined whether the category to which the image acquisition area belongs is a production site.

[0062] In some examples, it is possible to determine whether the corresponding area is a production site based on the aforementioned shooting location information, for example, by querying a preset shooting location information and area category comparison table. In some examples, it is also possible to determine the category to which the image acquisition area presented by the target image belongs with the help of a pre-trained neural network model. For example, the target image can be provided to the above model, and the model can output whether the category to which the image acquisition area belongs is a production site.

[0063] If the judgment at 520 is yes, at 522, the matching degree is adjusted upward according to the first score adjustment strategy to obtain the score of the image acquisition area under the 5S management standard.

[0064] In this embodiment, the first score adjustment strategy can be used to improve the score. In some examples, the first score adjustment strategy can adopt a linear adjustment method. For example, the matching degree can be multiplied by 1.1 to obtain the adjusted score. For another example, the adjusted score can be determined as (matching degree + 1) / 2. For example, if the matching degree is 0.75, the adjusted score is 0.75 × 1.1 = 0.825; or, the adjusted score is (0.75 + 1) ÷ 2 = 0.875. It can be understood that if the matching degree is a percentage system, the adjusted score can be determined as (matching degree + 100) / 2. In some examples, the first score adjustment strategy can be adjusted by a piecewise function. For example, the second preset threshold can be 70 (under a percentage system), when the matching degree is between 70 and 80, the adjustment range is an increase of 8%; when the matching degree is between 80 and 90, the adjustment range is an increase of 5%; when the matching degree is greater than 90, the adjustment range is an increase of 3% (not exceeding 100 points). It can be understood that the higher the matching degree, the better the on-site situation presented. Therefore, in the case of loose assessment, the above-mentioned segmented adjustment can be used to improve the score more finely according to the matching degree, so as to more clearly reflect the score difference in the scenario of better production site conditions.

[0065] If the determination result at 510 is no, at 530 , it is determined whether the matching degree is less than a second preset threshold.

[0066] In this embodiment, the second preset threshold can be determined according to the value range of the matching degree and the judgment standard for whether the office area has a poor on-site condition under the 5S management standard. In some examples, the value of the second preset threshold is generally adapted to the value range of the matching degree (e.g., 0~1, or 0~100). In some examples, the judgment standard for whether the office area has a poor on-site condition under the 5S management standard can be determined according to the score for the historical office area image, for example, the first place in the last 30% of the score of the historical office area image can be used as the score line to determine the second preset threshold. For example, there are 1,000 historical office area images in total, and the second preset threshold can be determined according to the score of the 701st historical office area image arranged from high to low according to the score. For example, if the score is 65 and the value range of the matching degree is 0~1, the second preset threshold can be determined as 6.5.

[0067] If the answer at step 530 is yes, at step 540 , it is determined whether the category to which the image acquisition area belongs is an office area.

[0068] In this embodiment, a method similar to step 520 may be used to determine whether the category to which the image acquisition area belongs is an office area. It should be noted that when the pre-trained neural network model can directly output the category to which the image acquisition area belongs (such as a production site, an office area, etc.), the model used in this step may be the same as step 520; when the pre-trained neural network model directly outputs whether the category to which the image acquisition area belongs is a certain category (such as a production site, an office area, etc.), for example, "1" indicates yes and "0" indicates no, the model used in this step is different from step 520.

[0069] If the result of the determination in step 540 is yes, then in step 542 , the matching degree is adjusted downward according to the second score adjustment strategy to obtain a score of the image acquisition area under the 5S management standard.

[0070] In this embodiment, the second score adjustment strategy is used to reduce the score. In some examples, the second score adjustment strategy can adopt a linear adjustment method. For example, a fixed constant can be subtracted from the matching degree to obtain an adjusted score. Among them, if the matching degree is a value between 0 and 1, the above-mentioned fixed constant can be, for example, 0.1; if the matching degree is a value between 0 and 100, the above-mentioned fixed constant can be, for example, 10. In some examples, the second score adjustment strategy can be adjusted by a piecewise function. For example, the second preset threshold value can be 0.7, and when the matching degree is between 0.6 and 0.7, the adjustment range is a reduction of 5%; when the matching degree is between 0.5 and 0.6, the adjustment range is a reduction of 8%; when the matching degree is less than 0.5, the adjustment range is a reduction of 10%. It can be understood that the lower the matching degree means the worse the on-site situation presented. Thus, in the case of strict assessment, the score difference in the scene of poor office area conditions can be more clearly reflected according to the above-mentioned segmented adjustment.

[0071] Since the 5S management standards, implementation difficulties and assessment strictness of different regions are different, they are usually related to the work scenes in the region. For example, for offices, the 5S management standards are relatively high, the implementation difficulty is relatively low, and the assessment is relatively strict; but for production sites with complex working conditions, if the same standards as offices are adopted, due to the high difficulty of implementation, it is possible to invest a lot of time and energy but still fail to achieve the desired results, and there is usually no need to adopt strict standards (such as requiring the ground and equipment surfaces to be spotless, etc.). It is only necessary to ensure that the necessary equipment and tools are placed according to the standards, check for safety hazards, and improve work efficiency. Therefore, the 5S management standards are relatively low and the assessment is relatively loose. In the above way, this solution can use the corresponding score adjustment strategy to deal with the difference between the actual 5S management situation of the image acquisition area and the standard situation and the category to which the image acquisition area belongs, which can reduce the large score differences caused by regional category factors, thereby facilitating the subsequent comparison between the various regions of the entire chemical enterprise, thereby promoting targeted improvements in 5S management.

[0072] Reference below Figure 6 , Figure 6 FIG. 6 is a signaling diagram showing an example of an interaction process 600 of a 5S management system according to an embodiment of the present specification. The process 600 may be Figure 3 The process 600 is applicable to the case where the feature description information includes multiple feature point descriptors.

[0073] like Figure 6 As shown, at 610, the image acquisition device 601 may adopt the shooting parameters used when shooting the reference image, shoot the image acquisition area from the shooting position, and obtain the target image.

[0074] In this embodiment, the shooting position and the shooting parameters used when shooting the reference image can be indicated by a shooting indicator set in the image acquisition area. The image acquisition device 601 can be of the same model as the device for shooting the reference image. The reference image can be an image of the image acquisition area when it meets the 5S management standard.

[0075] At 620 , the image acquisition device 601 may identify the region identification information corresponding to the target image.

[0076] In some examples, the image acquisition device 601 can identify the region identification information corresponding to the target image by scanning and parsing the region identification code in the form of a QR code. In some examples, the image acquisition device 601 can identify the region identification information corresponding to the target image by scanning and identifying an RFID tag. In some examples, after identifying the region identification information corresponding to the target image, the image acquisition device 601 can associate it with the target image.

[0077] At 630 , the image acquisition device 601 may send the target image and corresponding region identification information to the computing device 602 .

[0078] In some examples, the image acquisition device 601 can send the target image and the corresponding area identification information to the computing device 602 via a wired network (such as industrial Ethernet) or a wireless network (such as an enterprise's internal Wi-Fi network). In some examples, an encrypted transmission protocol can be used to ensure data security. In some examples, the image data can be compressed to reduce the amount of data transmission, improve transmission efficiency, and ensure that the image quality is within an acceptable range, so that the computing device can quickly receive and process the data.

[0079] At 640 , the computing device 602 may store the target image in a corresponding storage path with a corresponding name according to the region identification information and the predefined naming rule and storage rule.

[0080] In some examples, the predefined naming rule may be "area number_shooting time_image sequence number.jpg". For example, the first image taken of Area A of the raw material storage area at 10:30 on December 5, 2024 may be named "003_20241205103000_001.jpg". In some examples, the storage rules may stipulate that the storage path is divided according to the area category. The storage path of the above image may be, for example, " / Chemical Enterprise 5S Image Data / Raw Material Storage Area / Area A / ". The computing device 602 may store the image in a designated location with a corresponding name according to the above rules, providing a basis for subsequent data management and query. In some examples, the stored image may also be hashed to generate and save a unique hash value of the image for data integrity verification and to prevent the image from being tampered with.

[0081] At 650 , the computing device 602 may determine a reference image that matches the region identification information from a plurality of pre-stored reference images.

[0082] In this embodiment, the computing device 602 can maintain a reference image database. The reference image database can classify and store the reference images according to the area identification information. When the computing device 602 receives the area identification information sent by the image acquisition device 601, the corresponding reference image can be quickly retrieved through a database query statement. For example, the area number can be used as an index field in the database. When the area identification information of "003" is received, it can be determined that the reference image matching the area identification information is the reference image of area A of the raw material storage area.

[0083] At 660 , computing device 602 may determine a degree of match between the target image and the matching reference image.

[0084] At 670 , the computing device 602 may perform conversion in a corresponding manner according to the matching degree to obtain a score of the image acquisition area under the 5S management standard.

[0085] It should be noted that the above steps 610 and 620 can refer to Figure 2 The description of the photographing indicator 210 and the image acquisition device 220 in the embodiment. The above steps 660 and 670 can refer to Figure 2 Relevant description about step S201 and step S202 in the embodiment.

[0086] At 680 , the computing device 602 may determine a terminal that matches the area identification information from a pre-stored area responsibility relationship table.

[0087] In this embodiment, the regional responsibility relationship table can store the direct person in charge and the follow-up person in charge corresponding to each image acquisition area and the terminal information used by them. In some examples, the matching terminals may include a first terminal 603 and a second terminal 604. The first terminal 603 can be the terminal used by the direct person in charge corresponding to the regional identification information. The second terminal 604 can be the terminal used by the follow-up person in charge corresponding to the regional identification information. For example, the regional responsibility relationship table can record that the direct person in charge of the quality inspection area M of a chemical company is a quality inspector, and the follow-up person in charge can be a quality supervisor. The corresponding terminal information can be, for example, a mobile phone number, an email address, an account number of a company's office system, etc. After the computing device 602 derives the score of the quality inspection area M under the 5S management standard based on the target image of the quality inspection area M, the matching terminal information can be determined by querying the regional responsibility relationship table.

[0088] At 690 , the computing device 602 may send the score to the first terminal 603 .

[0089] In some examples, the computing device 602 can send the score to the first terminal 603 of the directly responsible person through the internal message push system of the enterprise. In some examples, the message push system can adopt a push mechanism similar to Apple's APNS (Apple Push Notification Service) or Android's FCM (Firebase Cloud Messaging) to ensure that the score information can be pushed to the terminal device of the directly responsible person in a timely and accurate manner. In some examples, the directly responsible person can be reminded to check the score information on the terminal device by means of pop-up windows, vibrations, or voice prompts. In some examples, the first terminal can display detailed score results, image comparisons (for example, showing the difference between the target image and the benchmark image in thumbnail form), and improvement suggestions in the corresponding application, so that the directly responsible person can understand the 5S management status of the area.

[0090] In some examples, the first terminal may be configured to create a 5S management task and save the corresponding task information in response to receiving a 5S management task creation instruction. In some examples, the 5S management task creation instruction may be triggered by a user clicking a "New Task" button on the page of the first terminal. The task information may include at least information related to the target image and the corresponding reference image. The reference image related information is used to determine the reference image corresponding to the target image, for example, the corresponding reference image itself, or the indication information for indicating the corresponding reference image, such as an area identifier, an image number, etc. Optionally, the task information may also include but is not limited to at least one of the following: creation date, direct person in charge, department, contact information of the person in charge of follow-up, follow-up situation, improvement measures, expected completion time, and remarks. The user may input the above task information by interacting with the first terminal. For example, the first terminal may be interacted with by text input, voice input, selection from a drop-down list, uploading the captured video, etc. In some examples, the first terminal may also have a near-expiry reminder function. For example, when there is still a certain length of time (for example, 2 days) from the expected completion time, the first terminal may automatically pop up a reminder window and / or emit a reminder sound to remind that the corresponding task is about to expire. In some examples, the first terminal may also be configured to manage the created 5S management tasks. For example, a task list may be generated according to the stored 5S management tasks and corresponding task information, and support clicking to view corresponding task details.

[0091] At 6100 , the computing device 602 sends a score that is lower than a preset threshold to the corresponding second terminal 604 .

[0092] In this embodiment, when the computing device determines that the score of a certain image acquisition area is lower than a preset threshold, the score can be sent to the second terminal 604 of the person responsible for follow-up. In some examples, the message push system can be used for sending. In some examples, the second terminal can highlight the score below the threshold, for example, the score, the area name, and prompt information such as "need to pay special attention and follow up for improvement" can be displayed in red font, so as to prompt the person responsible for follow-up to intervene in time, supervise and assist the person directly responsible in the rectification work of 5S management.

[0093] At 6110 , the computing device 602 may group and count the obtained scores according to the area identification information, and send the statistical results to each second terminal 604 .

[0094] In this embodiment, the computing device 602 can group and count the scores according to dimensions such as regional categories (such as production areas, storage areas, office areas, etc.), time periods (such as weekly, monthly, and quarterly), and send these statistical results to the second terminal 604 of the follow-up responsible person in each area in the form of charts (such as intuitive forms such as bar charts and line charts). For example, the average score, the highest score, the lowest score, and the score distribution of each production workshop can be counted once a month. For another example, the statistical information can also include the 5S management score completion rate determined based on the ratio of the number of areas that have completed the scoring to the number of areas that should be scored. Thus, the follow-up responsible person can view the 5S management statistical results of different areas in the corresponding application of the second terminal.

[0095] Through the above methods, this solution accurately collects images, calculates matching degrees and scores through efficient interaction between image acquisition equipment and computing equipment, and pushes the scores to relevant responsible persons in a timely manner. At the same time, it provides comprehensive statistical analysis results, which effectively improves the efficiency, quality and coverage of 5S management work, and helps to ensure the normalization and standardization of the production and operation environment of chemical enterprises.

[0096] use Figure 1-Figure 6 The 5S management system for chemical enterprises disclosed in the document can clarify the shooting position and shooting requirements through the photo indicators, and ensure the shooting consistency between the target image and the reference image captured by the image acquisition device; on this basis, the computing device can calculate the matching degree between the target image and the pre-stored reference image, and can further determine the score in an adaptive manner according to the size of the matching degree. This not only reduces the subjectivity of manual inspection, makes the scoring results more objective, and greatly improves the evaluation efficiency of 5S management, but also carefully distinguishes the differences in 5S management levels under different matching degrees, which is helpful to achieve refined management.

[0097] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0098] Not all steps and units in the above-mentioned processes and system structure diagrams are necessary, and some steps or units can be ignored according to actual needs. The execution order of each step is not fixed and can be determined according to needs. The device structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some units may be implemented by the same physical entity, or some units may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.

[0099] The term "exemplary" as used throughout this specification means "serving as an example, instance, or illustration" and does not imply "preferred" or "advantageous" over other embodiments. The detailed description includes specific details for the purpose of providing an understanding of the described techniques. However, the techniques may be implemented without these specific details. In some instances, in order to avoid obscuring the concepts of the described embodiments, well-known structures and devices are shown in block diagram form.

[0100] The optional implementation modes of the embodiments of the present specification are described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present specification are not limited to the specific details in the above implementation modes. Within the technical concept of the embodiments of the present specification, various simple modifications can be made to the technical solutions of the embodiments of the present specification, and these simple modifications all belong to the protection scope of the embodiments of the present specification.

[0101] The above description of the contents of this specification is provided to enable any person of ordinary skill in the art to implement or use the contents of this specification. Various modifications to the contents of this specification will be apparent to those of ordinary skill in the art, and the general principles defined herein may be applied to other variations without departing from the scope of protection of the contents of this specification. Therefore, the contents of this specification are not limited to the examples and designs described herein, but are consistent with the widest range of principles and novel features disclosed herein.

Claims

1. A 5S management system for a chemical enterprise, comprising: A photographing indicator disposed in the image acquisition area, wherein the photographing indicator is at least used to indicate a photographing position and photographing requirements; an image acquisition device configured to acquire a target image captured from the shooting position, for the image acquisition area, and meeting the shooting requirements; and A computing device is configured to determine a degree of match between the target image and a pre-stored reference image of the image acquisition area, wherein the reference image is an image of the image acquisition area when it complies with the 5S management standard; and convert the image in a corresponding manner according to the size of the degree of match to obtain a score of the image acquisition area under the 5S management standard.

2. The 5S management system according to claim 1, wherein: The computing device is further configured to: Acquiring feature description information of the reference image; Extracting the feature description information of the target image according to the generation method of the feature description information of the reference image; as well as A degree of match between the feature description information of the reference image and the feature description information of the target image is determined.

3. The 5S management system as claimed in claim 2, wherein: The feature description information includes a feature histogram, Determining the matching degree between the feature description information of the reference image and the feature description information of the target image includes: The matching degree is determined according to the similarity between the feature histogram of the reference image and the feature histogram of the target image.

4. The 5S management system as claimed in claim 2, wherein: The feature description information includes a plurality of feature point descriptors, wherein the feature point descriptors are used to describe local features of the feature points. Determining the matching degree between the feature description information of the reference image and the feature description information of the target image includes: Determining a similarity between a first number of feature point descriptors of the reference image and a second number of feature point descriptors of the target image; Determining the number of matching feature point pairs among the first number of feature points in the reference image and the second number of feature points in the target image according to the determined similarity; and The degree of matching between the feature description information of the reference image and the feature description information of the target image is determined according to at least one of the first number and the second number and the number of the matched feature point pairs.

5. The 5S management system according to claim 1, wherein: The conversion is performed in a corresponding manner according to the size of the matching degree to obtain the score of the image acquisition area under the 5S management standard, including: According to the size of the matching degree and the category to which the image acquisition area belongs, conversion is performed in a corresponding manner to obtain a score of the image acquisition area under the 5S management standard.

6. The 5S management system according to claim 5, wherein: The conversion is performed in a corresponding manner according to the size of the matching degree and the category to which the image acquisition area belongs, and the score of the image acquisition area under the 5S management standard is obtained, including: In response to the matching degree being greater than a first preset threshold and the category to which the image acquisition area belongs being a production site, the matching degree is adjusted upward according to a first score adjustment strategy to obtain a score of the image acquisition area under the 5S management standard, wherein the first score adjustment strategy is used to improve the score; and In response to the matching degree being less than a second preset threshold and the category to which the image acquisition area belongs being an office area, the matching degree is lowered according to a second score adjustment strategy to obtain a score of the image acquisition area under the 5S management standard, wherein the second score adjustment strategy is used to lower the score.

7. The 5S management system according to claim 1, wherein: The image acquisition device and the device for photographing the reference image have the same model, and the photographing requirements include photographing parameters used when photographing the reference image. The image acquisition device is configured to adopt the shooting parameters and shoot the image acquisition area from the shooting position to obtain the target image.

8. The 5S management system as claimed in claim 7, wherein: The photographing indicator includes an area identification code set in each image acquisition area, The image acquisition device is further configured to scan an area identification code set in an image acquisition area targeted by the target image, and identify area identification information corresponding to the target image.

9. The 5S management system according to claim 7, wherein: The image acquisition device supports a near field communication (NFC) function, and the photographing indicator includes a radio frequency identification (RFID) tag set at a designated photographing position in each image acquisition area. The image acquisition device also includes: an RFID tag identifier, The image acquisition device is further configured to, in response to the RFID tag identifier scanning the RFID tag, identify the area identification information and corresponding shooting parameters of the area indicated by the scanned RFID tag; when the current shooting parameters are consistent with the identified shooting parameters, shoot the image acquisition area to obtain the target image; when the current shooting parameters are inconsistent with the identified shooting parameters or the RFID tag is not scanned, shooting is not allowed.

10. The 5S management system according to claim 8 or 9, wherein: The computing device is further configured to: According to the area identification information, the target image is stored in a corresponding storage path with a corresponding name according to a predefined naming rule and storage rule; Determining a reference image that matches the area identification information from a plurality of pre-stored reference images; Determine a terminal that matches the regional identification information from a pre-stored regional responsibility relationship table, wherein the matching terminal includes a first terminal and a second terminal, wherein the first terminal is a terminal used by a person directly responsible for the regional identification information, and the second terminal is a terminal used by a person responsible for following up the regional identification information; sending the score to the first terminal; Sending scores below a preset threshold to a corresponding second terminal; as well as The obtained scores are grouped and counted according to the area identification information, and the statistical results are sent to each second terminal.

Images