A warehousing identification method, device and equipment of an electric power tool and a storage medium

Abstract

The application discloses a warehouse entry identification method, device and equipment of electric power tools and a storage medium. The method comprises the following steps: when the electric power tools enter the warehouse, obtaining an image to be identified of the electric power tools, performing feature point processing on the image to be identified through a three-dimensional feature point algorithm to obtain a feedback image, calculating a matching score of the feedback image and a product image of each electric power tool to be entered into the warehouse, and updating the state of the electric power tool to be entered into the warehouse when the matching score of the electric power tool to be entered into the warehouse is not less than a preset threshold score. It can be seen that no matter what kind of electric power tool enters the warehouse, the image to be identified can be obtained for identification and analysis when entering the warehouse, the feature matching of the image to be identified and the product image in the warehouse file is performed, and the parameter change of the electric power tool to be entered into the warehouse is completed when the matching is successful. The warehouse entry work of the electric power tool is realized through the image recognition mode, the loss of the additional auxiliary warehouse entry mark is avoided, the maintenance cost is reduced, and efficient warehouse entry is realized.

Description

Technical Field

[0001] This application relates to the field of intelligent identification, and more specifically, to a method, device, equipment, and storage medium for identifying electrical tools upon entry into a warehouse. Background Technology

[0002] With the development of power technology, the demand for power construction and maintenance has increased, requiring power technicians to work on-site. At power construction sites, various abandoned power tools are often found, and after construction, all tools need to be collected and inventoried. This process of collecting and inventorying tools is time-consuming, labor-intensive, and prone to errors, resulting in limited effectiveness. To address this, for the warehousing of power tools, technicians have implemented a data transmission method using electronic tags attached to the tools. This allows the system to scan the tags upon entry into the warehouse to retrieve the tool's information.

[0003] However, this method of attaching electronic tags to electrical tools is prone to damage during electrical operations, has high maintenance costs, and electronic tags are not universally applicable to all electrical tools; some wires and cables also cannot be used to attach electronic tags for extended periods.

[0004] Therefore, how to achieve low-damage and efficient storage of electrical tools is an issue that needs attention. Summary of the Invention

[0005] In view of the above problems, this application is made to provide a method, apparatus, equipment and storage medium for identifying electrical tools upon entry into the warehouse, so as to achieve low-damage and efficient entry of electrical tools into the warehouse.

[0006] To achieve the above objectives, the following specific solutions are proposed:

[0007] A method for identifying electrical tools upon entry into a warehouse, comprising:

[0008] When electrical tools enter the warehouse, an image of the electrical tools to be identified is acquired;

[0009] The image to be identified is processed by a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0010] For each electrical appliance to be put into storage in the warehouse archives, calculate the matching score of the feature matching between the feedback image and the product image, and use it as the matching score of the electrical appliance to be put into storage.

[0011] When it is detected that the matching score of any of the power tools to be added to the warehouse is not less than the preset threshold score, the status of the power tools to be added to the warehouse is updated to "added to the warehouse".

[0012] Optionally, the method further includes:

[0013] When the matching score of any of the power tools to be added to the database is less than the preset threshold score, a prompt message is sent to the management terminal to remind the user to add the power tools.

[0014] Optionally, the step of processing the image to be identified using a three-dimensional feature point algorithm to obtain the feedback image of the power tool includes:

[0015] The contour feature curve of the image to be identified is extracted using an image contour feature raster, and several feature points on the contour feature curve are determined.

[0016] Three feature points are randomly selected from the plurality of feature points, and one feature point is selected from the three feature points as the intersection point of two straight lines. A first straight line is constructed by connecting the intersection point with a first connection point, and a second straight line is constructed by connecting the intersection point with a second connection point. The first connection point and the second connection point are both non-intersecting points among the three feature points, and the first connection point and the second connection point are different.

[0017] When the angle between the first straight line and the second straight line is greater than a preset angle threshold, the contour feature curve is divided into several curve segments;

[0018] Calculate the first and second fractal dimensions of each curve segment;

[0019] When the first fractal dimension of two adjacent curve segments is the same, and the second fractal dimension of the two curve segments is the same, the two curve segments are merged to obtain the feedback image of the power tool.

[0020] Optionally, dividing the contour feature curve into several curve segments includes:

[0021] Using the feature points on the contour feature curve as segment boundary points, the contour feature curve is divided into several curve segments, each curve segment containing at least two feature points.

[0022] Optionally, calculating the first fractal dimension and the second fractal dimension of each curve segment includes:

[0023] Calculate the first distance from each feature point on each curve segment to the first straight line;

[0024] Calculate the second distance from each feature point on each curve segment to the second straight line;

[0025] Calculate the average of the first distances of each feature point on each curve segment, and determine it as the first average error of that curve segment;

[0026] Calculate the average of the second distances of each feature point on each curve segment, and determine it as the second average error of that curve segment;

[0027] The first fractal dimension of the curve segment is calculated based on the first average error of each curve segment and the number of curve segments of the contour feature curve.

[0028] The second fractal dimension of the curve segment is calculated based on the second average error of each curve segment and the number of curve segments of the contour feature curve.

[0029] Optionally, the method further includes:

[0030] When the angle between the first line and the second line is not greater than the preset angle threshold, the process returns to the step of arbitrarily selecting three feature points from the plurality of feature points, and selecting one feature point from the three feature points as the intersection point of the two lines, constructing a first line by connecting the intersection point with the first connection point, and a second line by connecting the intersection point with the second connection point, to update the first line and the second line until the angle between the updated first line and the updated second line is greater than the preset angle threshold, at which point the updated first line is determined to be the latest first line and the updated second line is determined to be the latest second line.

[0031] Optionally, for each product image of a power tool to be received into the warehouse archives, a matching score is calculated between the feedback image and the product image based on feature matching, including:

[0032] For each feature point of the contour feature curve in the feedback image, a target feature point matching that feature point is determined in the product image of each electrical appliance to be put into the warehouse in the archives of the warehouse.

[0033] For each feature point of the contour feature curve in the feedback image, the matching sub-score between the feature point and the target feature point that matches the feature point is calculated using the following formula, and is used as the matching sub-score of the feature point:

[0034]

[0035] Where, a is the feature vector corresponding to a feature point of the contour feature curve in the feedback image, b is the feature vector corresponding to the feature point in the product image that matches feature point a, (x a y a (x) represents the feature point corresponding to the feature vector a. b y b ) represents the feature point corresponding to the feature vector b;

[0036] The matching sub-scores of each feature point of the contour feature curve in the feedback image are summed to obtain the total matching score of the contour feature curve in the feedback image.

[0037] Divide the total matching score by the number of feature points in the product image of each power tool to be put into storage to obtain the matching score of the feature matching between the feedback image and the product image of the power tool to be put into storage.

[0038] A warehouse entry identification device for electrical tools, comprising:

[0039] An image acquisition unit is used to acquire an image of the electrical tools to be identified when the electrical tools enter the warehouse.

[0040] The feature point processing unit is used to process the image to be identified using a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0041] The matching score calculation unit is used to calculate the matching score of the feature matching between the feedback image and the product image for each power tool to be put into the warehouse in the archives of the warehouse, and use it as the matching score of the power tool to be put into the warehouse.

[0042] The inventory update unit is used to update the status of any electrical appliance to be entered into the inventory to "already entered into the inventory" when it is detected that the matching score of any electrical appliance to be entered into the inventory is not less than a preset threshold score.

[0043] Optionally, the device may also include:

[0044] The prompting unit is used to send a prompt message to the management terminal to remind the user to supplement the list of electrical tools when the matching score of any of the electrical tools to be added to the database is less than a preset threshold score.

[0045] Optionally, the feature point processing unit includes:

[0046] The first feature point processing subunit is used to extract the contour feature curve of the image to be identified through the image contour feature grid, and determine a number of feature points on the contour feature curve.

[0047] The second feature point processing subunit is used to arbitrarily select three feature points from the plurality of feature points, and select one feature point from the three feature points as the intersection point of two straight lines, construct a first straight line obtained by connecting the intersection point with a first connection point, and a second straight line obtained by connecting the intersection point with a second connection point. The first connection point and the second connection point are both non-intersecting points among the three feature points, and the first connection point and the second connection point are different.

[0048] The third feature point processing subunit is used to divide the contour feature curve into several curve segments when the angle between the first straight line and the second straight line is greater than a preset angle threshold.

[0049] The fourth feature point processing subunit is used to calculate the first fractal dimension and the second fractal dimension of each curve segment;

[0050] The fifth feature point processing subunit is used to merge two curve segments when the first fractal dimension of two adjacent curve segments is the same and the second fractal dimension of the two curve segments is the same, so as to obtain the feedback image of the power tool.

[0051] Optionally, the third feature point processing subunit includes:

[0052] The curve segment division unit is used to divide the contour feature curve into several curve segments, with feature points on the contour feature curve as segment boundary points, when the angle between the first straight line and the second straight line is greater than a preset angle threshold. Each curve segment contains at least two feature points.

[0053] Optionally, the fourth feature point processing subunit includes:

[0054] The first distance calculation unit is used to calculate the first distance from each feature point on each curve segment to the first straight line;

[0055] The second distance calculation unit is used to calculate the second distance from each feature point on each curve segment to the second straight line;

[0056] The first average error calculation unit is used to calculate the average value of the first distance of each feature point on each curve segment and determine it as the first average error of the curve segment.

[0057] The second average error calculation unit is used to calculate the average value of the second distance of each feature point on each curve segment and determine it as the second average error of the curve segment.

[0058] The first fractal dimension calculation unit is used to calculate the first fractal dimension of the curve segment based on the first average error of each curve segment and the number of curve segments of the contour feature curve.

[0059] The second fractal dimension calculation unit is used to calculate the second fractal dimension of the curve segment based on the second average error of each curve segment and the number of curve segments of the contour feature curve.

[0060] Optionally, the feature point processing unit further includes:

[0061] The sixth feature point processing subunit is used to, when the angle between the first line and the second line is not greater than the preset angle threshold, return to execute the steps of arbitrarily selecting three feature points from the plurality of feature points, selecting one feature point from the three feature points as the intersection point of the two lines, constructing a first line obtained by connecting the intersection point with a first connection point, and a second line obtained by connecting the intersection point with a second connection point, so as to update the first line and the second line until the angle between the updated first line and the updated second line is greater than the preset angle threshold, and then determine that the updated first line is the latest first line and the updated second line is the latest second line.

[0062] Optionally, the matching score calculation unit includes:

[0063] The first matching score calculation subunit is used to determine, for each feature point of the contour feature curve in the feedback image, the target feature point that matches the feature point in the product image of each electrical appliance to be put into the warehouse in the archives.

[0064] The second matching score calculation subunit is used to calculate, for each feature point of the contour feature curve in the feedback image, the matching sub-score between the feature point and the target feature point that matches the feature point using the following formula, and use it as the matching sub-score of the feature point:

[0065]

[0066] Where, a is the feature vector corresponding to a feature point of the contour feature curve in the feedback image, b is the feature vector corresponding to the feature point in the product image that matches feature point a, (x a y a (x) represents the feature point corresponding to the feature vector a. b y b ) represents the feature point corresponding to the feature vector b;

[0067] The third matching score calculation subunit is used to sum the matching sub-scores of each feature point of the contour feature curve in the feedback image to obtain the total matching score of the contour feature curve in the feedback image.

[0068] The fourth matching score calculation subunit is used to divide the total matching score by the number of feature points in the product image of each power tool to be put into storage, to obtain the matching score of the feature matching between the feedback image and the product image of the power tool to be put into storage, and use it as the matching score of the power tool to be put into storage.

[0069] A device for identifying electrical tools upon entry into a warehouse, comprising a memory and a processor;

[0070] The memory is used to store programs;

[0071] The processor is used to execute the program to implement the various steps of the electrical equipment entry identification method as described above.

[0072] A storage medium storing a computer program, which, when executed by a processor, implements the various steps of the warehousing identification method for electrical tools as described above.

[0073] By employing the above technical solution, this application acquires an image of the electrical tool to be identified when it enters the warehouse. A three-dimensional feature point algorithm is used to process the feature points of the image to obtain a feedback image of the electrical tool. For each electrical tool to be entered into the warehouse, a matching score is calculated between the feedback image and the product image, and this score is used as the matching score for that electrical tool. When the matching score of any electrical tool is detected to be not less than a preset threshold score, the status of that electrical tool is updated to "entered into the warehouse." Therefore, regardless of the type of electrical tool entering the warehouse, the application can perform identification analysis based on the image to be identified, matching it with the product image of the electrical tool in the warehouse archive. Upon successful matching, the parameters of the electrical tool are changed. This image recognition method enables the entry of electrical tools into the warehouse, avoiding the loss of additional auxiliary entry markers, reducing maintenance costs, and enabling efficient entry. Attached Figure Description

[0074] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0075] Figure 1 A schematic diagram illustrating the process of identifying electrical tools upon entry into a warehouse, provided as an embodiment of this application;

[0076] Figure 2 A schematic diagram of a device for identifying the entry of electrical tools into a warehouse, provided in an embodiment of this application;

[0077] Figure 3 This is a schematic diagram of the structure of a device for identifying the entry of electrical tools into a warehouse, provided as an embodiment of this application. Detailed Implementation

[0078] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0079] The proposed solution can be implemented based on a terminal with data processing capabilities, such as a computer, server, or cloud platform.

[0080] Next, combined Figure 1 The method for identifying electrical tools entering the warehouse according to this application may include the following steps:

[0081] Step S110: When electrical tools enter the warehouse, obtain the image of the electrical tools to be identified.

[0082] Specifically, the image to be identified for the electrical equipment can represent a surface image captured when the electrical equipment enters the warehouse. When the electrical equipment enters the warehouse, it can be photographed by an image acquisition camera to obtain a surface image of the electrical equipment, which can be transmitted to the terminal in real time.

[0083] Step S120: Perform feature point processing on the image to be identified using a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0084] Specifically, the three-dimensional feature point algorithm can be pre-stored in the terminal. When an image of an electrical tool to be identified is obtained, the three-dimensional feature point algorithm can be retrieved and applied to the image to be identified.

[0085] The algorithm for the three-dimensional feature point algorithm can be described as follows: the image to be identified is identified by using an image contour feature grid with logarithmic polar coordinates; the identified feature points are then analyzed by segmented contours; finally, some segments are selected and merged to process the image to be identified and obtain a feedback image.

[0086] It is understandable that the feedback image is obtained by processing the image to be recognized, therefore the coordinates of the feature points in the feedback image are not exactly the same as the coordinates of the feature points in the image to be recognized.

[0087] Step S130: For each product image of a power tool to be put into storage in the warehouse archive, calculate the matching score of the feature matching between the feedback image and the product image, and use it as the matching score of the power tool to be put into storage.

[0088] Specifically, the matching score can represent the degree of matching between the feedback image and the product image.

[0089] In this case, the number of the same type of electrical tool recorded in the warehouse file can be more than one. Therefore, when a type of electrical tool is not completely stored in the warehouse, its status can be displayed as "pending storage," and the number of items pending storage can be indicated. "Pending storage" indicates that the electrical tool of that type is not completely stored. For example, if the number of type A electrical tools recorded in the warehouse file is 3, then when only 2 of these electrical tools have been stored, the electrical tool of that type will still be in the "pending storage" status.

[0090] Understandably, the warehouse records the details of various electrical tools entering the warehouse in real time. When a type of electrical tool to be entered into the warehouse is fully entered, its status can be updated to "entered into the warehouse". Therefore, whenever an electrical tool enters the warehouse, the object to be matched / compared is the product image of each electrical tool that has not yet been entered into the warehouse. The matching degree of the images is analyzed and evaluated from the product images of each electrical tool to be entered into the warehouse.

[0091] Step S140: When it is detected that the matching score of any of the power tools to be put into storage is not less than the preset threshold score, the power tool is put into storage operation and the storage quantity of the power tool to be put into storage is updated.

[0092] Specifically, the preset threshold score can represent the minimum score required for a successful match between a product image and a feedback image. If the matching score of any power tool to be put into storage is not less than the preset threshold score, it indicates that the product image of the power tool to be put into storage has successfully matched the feedback image of the power tool to be put into storage. This means that the power tool to be put into storage is the power tool to be put into storage recorded in the warehouse file. In this case, the power tool can be put into storage operation, and the quantity of the power tool to be put into storage can be updated.

[0093] The warehousing identification method for electrical tools provided in this embodiment acquires an image of the electrical tool to be identified when it enters the warehouse. A three-dimensional feature point algorithm is used to process the feature points of the image to obtain a feedback image of the electrical tool. For each electrical tool to be warehoused in the warehouse archive, a matching score is calculated between the feedback image and the product image, and this score is used as the matching score for that electrical tool. When the matching score of any electrical tool to be warehoused is detected to be not less than a preset threshold score, the status of that electrical tool is updated to "warehousing complete". Therefore, regardless of the type of electrical tool entering the warehouse, the method can identify and analyze the image to be identified during the warehousing process, matching its features with the product images of other electrical tools in the warehouse archive. Upon successful matching, the parameters of the electrical tool to be warehoused are changed. This image recognition method enables the warehousing of electrical tools, avoiding the loss of additional auxiliary warehousing markers, reducing maintenance costs, and enabling efficient warehousing.

[0094] In some embodiments of this application, considering the possibility that the comparison between the feedback image and the product image fails, it is likely due to the fact that the electrical tool has not been pre-entered into the warehouse archives. Therefore, the administrator can be notified for intervention. Based on this, the method for identifying the entry of electrical tools into the warehouse provided by this application may further include:

[0095] When the matching score of any power tool to be put into storage is less than the preset threshold score, a prompt message is sent to the management terminal to remind the administrator to supplement the power tool. This is to remind the administrator of the control terminal that no matching power tool is found in the warehouse file and manual intervention is required.

[0096] Understandably, administrators can add product images of new products to the warehouse archives, enabling subsequent identification of similar electrical tools to successfully match features and smoothly execute the warehouse entry process.

[0097] In some embodiments of this application, the process of performing feature point processing on the image to be identified using a three-dimensional feature point algorithm to obtain the feedback image of the power tool is described. This process may include:

[0098] S1. Extract the contour feature curve of the image to be identified through the image contour feature raster, and determine several feature points on the contour feature curve.

[0099] Specifically, the image contour feature grid is covered with coordinate points, which can be used to extract the contour feature curve of the image to be recognized and describe the feature points of the contour curve of the image to be recognized.

[0100] The number of feature points on the contour feature curve can be determined based on the size and structural characteristics of the electrical tools, or the number of feature points on the contour feature curve can be customized.

[0101] S2. Select any three feature points from a number of feature points, and choose one of these three feature points as the intersection point of two straight lines. Construct a first straight line by connecting the intersection point with the first connection point, and a second straight line by connecting the intersection point with the second connection point.

[0102] Among them, the first connection point and the second connection point are both non-intersecting points among the three feature points, and the first connection point is different from the second connection point.

[0103] Specifically, the method for selecting the intersection point from the three feature points can be random selection.

[0104] For example, if three feature points are selected from a number of feature points, namely A, B and C, and B is selected as the intersection point, then A and C can be determined as the first connection point and the second connection point respectively (or C is the first connection point and A is the second connection point). Thus, connecting A and B, and C and B respectively, we obtain the first line AB and the second line AC (or AC is the first line and AB is the second line).

[0105] S3. When the angle between the first straight line and the second straight line is greater than the preset angle threshold, the contour feature curve is divided into several curve segments.

[0106] The preset angle threshold represents the minimum angle between the first and second lines when they can serve as valid reference lines. The preset angle threshold can be customized.

[0107] It is understandable that when the angle between the first straight line and the second straight line is greater than the preset angle threshold, it can be said that the first straight line and the second straight line can be used as effective reference lines for analyzing the contour feature curve of the image to be identified of the power tool. Then the contour feature curve analysis can continue and the contour feature curve can be divided into several curve segments.

[0108] Specifically, the process of dividing the contour feature curve into several curve segments can be as follows:

[0109] Using the feature points on the contour feature curve as segment boundary points, the contour feature curve is divided into several curve segments, each curve segment containing at least two feature points.

[0110] The number of segments in the contour feature curve can be customized.

[0111] Furthermore, when the angle between the first straight line and the second straight line is not greater than a preset angle threshold, the process of returning to execution S2 of this embodiment, arbitrarily selecting three feature points from a number of feature points, selecting one feature point from these three feature points as the intersection point of the two straight lines, constructing the first straight line obtained by connecting the intersection point with the first connection point, and the second straight line obtained by connecting the intersection point with the second connection point.

[0112] It is understandable that when the angle between the first line and the second line is not greater than a preset angle threshold, it means that the first line and the second line cannot be used as valid reference lines for analyzing the contour feature curve of the image to be identified of the power tool. In this case, the first line and the second line need to be reselected, that is, the S2 step of this embodiment needs to be returned to update the first line and the second line until the angle between the updated first line and the updated second line is greater than the preset angle threshold. Then, the updated first line is determined to be the latest first line and the updated second line is determined to be the latest second line.

[0113] S4. Calculate the first fractal dimension and the second fractal dimension of each curve segment.

[0114] Specifically, the first fractal dimension of each curve segment can represent the degree of error deviation of each feature point on the curve segment with respect to the first straight line, and the second fractal dimension of each curve segment can represent the degree of error deviation of each feature point on the curve segment with respect to the second straight line.

[0115] The process of calculating the first and second fractal dimensions of each curve segment can include:

[0116] S41. Calculate the first distance from each feature point on each curve segment to the first straight line.

[0117] Specifically, the first distance from each feature point on each curve segment to the first straight line can be calculated using the following formula:

[0118]

[0119] Where d1 represents the distance from the feature point (x1, y1) to the first line: R1x+T1y+S1=0.

[0120] S42. Calculate the second distance from each feature point on each curve segment to the second straight line.

[0121] Specifically, the first distance from each feature point on each curve segment to the second straight line can be calculated using the following formula:

[0122]

[0123] Where d2 represents the distance from the feature point (x2, y2) to the second line: R1x+T2y+S2=0.

[0124] S43. Calculate the average of the first distances of each feature point on each curve segment, and determine it as the first average error of the curve segment.

[0125] Specifically, the average first distance of each feature point on each curve segment can be calculated using the following formula:

[0126] δ1=(d 11 +d 12 +d 13 +...+d 1n ) / n

[0127] Where, d 11 d 12 d 13 , ..., d 1n Both are the first distances between feature points on each curve segment, and n is the number of feature points on that curve segment.

[0128] S44. Calculate the average of the second distances of each feature point on each curve segment, and determine it as the second average error of the curve segment.

[0129] Specifically, the average second distance of each feature point on each curve segment can be calculated using the following formula:

[0130] δ2=(d 21 +d 22 +d 23 +...+d 2n ) / n

[0131] Where, d 21 d 22 d 23 , ..., d 2n Both are the second distances to the feature points on each curve segment, and n is the number of feature points on that curve segment.

[0132] S45. Calculate the first fractal dimension of each curve segment based on the first average error of each curve segment and the number of curve segments of the contour feature curve.

[0133] Specifically, the first fractal dimension of each curve segment can be calculated using the following formula:

[0134]

[0135] Where δ1 is the first average error of each curve segment, and N is the number of curve segments of the contour feature curve.

[0136] S46. Calculate the second fractal dimension of each curve segment based on the second average error of each curve segment and the number of curve segments of the contour feature curve.

[0137] Specifically, the second fractal dimension of each curve segment can be calculated using the following formula:

[0138]

[0139] Where δ2 is the second average error of each curve segment, and N is the number of curve segments of the contour feature curve.

[0140] S5. When the first fractal dimension of two adjacent curve segments is the same, and the second fractal dimension of the two curve segments is the same, merge the two curve segments to obtain the feedback image of the power tool.

[0141] It is understandable that after merging two adjacent curve segments, the position / coordinates of the feature points on the resulting feedback image have changed. Therefore, the position / coordinates of the feature points on the feedback image are different from the position / coordinates of the feature points on the image to be identified before merging.

[0142] In some embodiments of this application, the process of calculating the matching score of the feedback image and the product image for each electrical appliance to be put into storage in the warehouse archives mentioned in the above embodiments is described. This process may include:

[0143] S1. For each feature point of the contour feature curve in the feedback image, determine the target feature point that matches the feature point in the product image of each electrical appliance to be put into the warehouse in the warehouse archive.

[0144] It is important to note that there may be more feature points in the product image than on the contour feature curve in the feedback image. Therefore, not every feature point in the product image can be matched with a feature point on the contour feature curve in the feedback image. In other words, not every feature point in the product image can be used as a target feature point. Only a feature point that successfully matches a feature point on the contour feature curve in the feedback image is considered a target feature point.

[0145] S2. For each feature point of the contour feature curve in the feedback image, calculate the matching sub-score between the feature point and the target feature point that matches the feature point using the following formula, and use it as the matching sub-score of the feature point:

[0146]

[0147] Where, a is the feature vector corresponding to a feature point of the contour feature curve in the feedback image, b is the feature vector corresponding to the feature point that matches feature point a in the product image, (xa y a (x) represents the feature point corresponding to the feature vector a. b y b ) represents the feature point corresponding to the feature vector b.

[0148] Specifically, the matching sub-score S can represent the cosine of the angle between feature vector a and feature vector b, and its value is no greater than 1.

[0149] S3. Sum the matching sub-scores of each feature point of the contour feature curve in the feedback image to obtain the total matching score of the contour feature curve in the feedback image.

[0150] It is understandable that the total matching score is the sum of the matching sub-scores of each feature point on the contour feature curve in the feedback image, which can reflect the number of feature matches between the feedback image and the product image.

[0151] S4. Divide the total matching score by the number of feature points in the product image of each power tool to be put into storage to obtain the matching score of the feature matching between the feedback image and the product image of the power tool to be put into storage.

[0152] It is understandable that the numerator of the matching score is the sum of the matching sub-scores of each feature point on the contour feature curve in the feedback image, and the denominator of the matching score is the number of feature points in the product image. Therefore, the matching score can reflect the overall degree of matching between the feedback image and the product image.

[0153] For example, if a product image has many feature points, and the feature points on the contour feature curve of the feedback image are not successfully matched, the numerator of the matching score is fixed and the denominator is large, so the matching score will be low.

[0154] The following describes the apparatus for identifying electrical tools entering the warehouse according to the embodiments of this application. The apparatus for identifying electrical tools entering the warehouse described below can be referred to in correspondence with the method for identifying electrical tools entering the warehouse described above.

[0155] See Figure 2 , Figure 2 This is a schematic diagram of a device for identifying electrical tools upon entry into a warehouse, as disclosed in an embodiment of this application.

[0156] like Figure 2 As shown, the device may include:

[0157] Image acquisition unit 11 is used to acquire an image of the electrical tools to be identified when the electrical tools enter the warehouse;

[0158] The feature point processing unit 12 is used to process the feature points of the image to be identified using a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0159] The matching score calculation unit 13 is used to calculate the matching score of the feature matching between the feedback image and the product image for each power tool to be put into the warehouse in the archives of the warehouse, and use it as the matching score of the power tool to be put into the warehouse.

[0160] The inventory update unit 14 is used to update the status of any electrical appliance to be entered into the inventory to "already entered into the inventory" when it is detected that the matching score of any electrical appliance to be entered into the inventory is not less than a preset threshold score.

[0161] Optionally, the device may also include:

[0162] The prompting unit is used to send a prompt message to the management terminal to remind the user to supplement the list of electrical tools when the matching score of any of the electrical tools to be added to the database is less than a preset threshold score.

[0163] Optionally, the feature point processing unit includes:

[0164] The first feature point processing subunit is used to extract the contour feature curve of the image to be identified through the image contour feature grid, and determine a number of feature points on the contour feature curve.

[0165] The second feature point processing subunit is used to arbitrarily select three feature points from the plurality of feature points, and select one feature point from the three feature points as the intersection point of two straight lines, construct a first straight line obtained by connecting the intersection point with a first connection point, and a second straight line obtained by connecting the intersection point with a second connection point. The first connection point and the second connection point are both non-intersecting points among the three feature points, and the first connection point and the second connection point are different.

[0166] The third feature point processing subunit is used to divide the contour feature curve into several curve segments when the angle between the first straight line and the second straight line is greater than a preset angle threshold.

[0167] The fourth feature point processing subunit is used to calculate the first fractal dimension and the second fractal dimension of each curve segment;

[0168] The fifth feature point processing subunit is used to merge two curve segments when the first fractal dimension of two adjacent curve segments is the same and the second fractal dimension of the two curve segments is the same, so as to obtain the feedback image of the power tool.

[0169] Optionally, the third feature point processing subunit includes:

[0170] The curve segment division unit is used to divide the contour feature curve into several curve segments, with feature points on the contour feature curve as segment boundary points, when the angle between the first straight line and the second straight line is greater than a preset angle threshold. Each curve segment contains at least two feature points.

[0171] Optionally, the fourth feature point processing subunit includes:

[0172] The first distance calculation unit is used to calculate the first distance from each feature point on each curve segment to the first straight line;

[0173] The second distance calculation unit is used to calculate the second distance from each feature point on each curve segment to the second straight line;

[0174] The first average error calculation unit is used to calculate the average value of the first distance of each feature point on each curve segment and determine it as the first average error of the curve segment.

[0175] The second average error calculation unit is used to calculate the average value of the second distance of each feature point on each curve segment and determine it as the second average error of the curve segment.

[0176] The first fractal dimension calculation unit is used to calculate the first fractal dimension of the curve segment based on the first average error of each curve segment and the number of curve segments of the contour feature curve.

[0177] The second fractal dimension calculation unit is used to calculate the second fractal dimension of the curve segment based on the second average error of each curve segment and the number of curve segments of the contour feature curve.

[0178] Optionally, the feature point processing unit further includes:

[0179] The sixth feature point processing subunit is used to, when the angle between the first line and the second line is not greater than the preset angle threshold, return to execute the steps of arbitrarily selecting three feature points from the plurality of feature points, selecting one feature point from the three feature points as the intersection point of the two lines, constructing a first line obtained by connecting the intersection point with a first connection point, and a second line obtained by connecting the intersection point with a second connection point, so as to update the first line and the second line until the angle between the updated first line and the updated second line is greater than the preset angle threshold, and then determine that the updated first line is the latest first line and the updated second line is the latest second line.

[0180] Optionally, the matching score calculation unit includes:

[0181] The first matching score calculation subunit is used to determine, for each feature point of the contour feature curve in the feedback image, the target feature point that matches the feature point in the product image of each electrical appliance to be put into the warehouse in the archives.

[0182] The second matching score calculation subunit is used to calculate, for each feature point of the contour feature curve in the feedback image, the matching sub-score between the feature point and the target feature point that matches the feature point using the following formula, and use it as the matching sub-score of the feature point:

[0183]

[0184] Where, a is the feature vector corresponding to a feature point of the contour feature curve in the feedback image, b is the feature vector corresponding to the feature point in the product image that matches feature point a, (x a y a (x) represents the feature point corresponding to the feature vector a. b y b ) represents the feature point corresponding to the feature vector b;

[0185] The third matching score calculation subunit is used to sum the matching sub-scores of each feature point of the contour feature curve in the feedback image to obtain the total matching score of the contour feature curve in the feedback image.

[0186] The fourth matching score calculation subunit is used to divide the total matching score by the number of feature points in the product image of each power tool to be put into storage, to obtain the matching score of the feature matching between the feedback image and the product image of the power tool to be put into storage, and use it as the matching score of the power tool to be put into storage.

[0187] The device for identifying the entry of electrical tools provided in this application embodiment can be applied to devices for identifying the entry of electrical tools, such as terminals like mobile phones and computers. Optionally, Figure 3 This diagram shows the hardware structure of the device for identifying electrical tools upon entry into the warehouse. (Refer to...) Figure 3 The hardware structure of the device for identifying the entry of electrical tools into the warehouse may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4.

[0188] In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4;

[0189] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0190] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0191] The memory stores a program, which the processor can call. The program is used for:

[0192] When electrical tools enter the warehouse, an image of the electrical tools to be identified is acquired;

[0193] The image to be identified is processed by a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0194] For each electrical appliance to be put into storage in the warehouse archives, calculate the matching score of the feature matching between the feedback image and the product image, and use it as the matching score of the electrical appliance to be put into storage.

[0195] When it is detected that the matching score of any of the power tools to be put into storage is not less than the preset threshold score, the storage operation of the power tools is completed and the storage quantity of the power tools to be put into storage is updated.

[0196] Optionally, the refined and extended functions of the program can be found in the description above.

[0197] This application embodiment also provides a storage medium that can store a program suitable for execution by a processor, the program being used for:

[0198] When electrical tools enter the warehouse, an image of the electrical tools to be identified is acquired;

[0199] The image to be identified is processed by a three-dimensional feature point algorithm to obtain the feedback image of the power tool.

[0200] For each electrical appliance to be put into storage in the warehouse archives, calculate the matching score of the feature matching between the feedback image and the product image, and use it as the matching score of the electrical appliance to be put into storage.

[0201] When it is detected that the matching score of any of the power tools to be put into storage is not less than the preset threshold score, the storage operation of the power tools is completed and the storage quantity of the power tools to be put into storage is updated.

[0202] Optionally, the refined and extended functions of the program can be found in the description above.

[0203] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0204] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0205] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A warehouse entry identification method for a power tool, characterized by, The method comprises the following steps: When the electric power tool enters the warehouse, an image to be recognized of the electric power tool is acquired; A feature point processing is performed on the image to be recognized by a three-dimensional feature point algorithm to obtain a feedback image of the electric power tool; For a product image of each electric power tool to be stored in the archive of the warehouse, a matching score of the feedback image and the product image is calculated, and the matching score is used as the matching score of the electric power tool to be stored; When it is detected that the matching score of any electric power tool to be stored is not less than a preset threshold score, the electric power tool entering the warehouse is completed, and the number of the electric power tool to be stored is updated; The feature point processing on the image to be recognized by the three-dimensional feature point algorithm to obtain the feedback image of the electric power tool comprises the following steps: An outline feature curve of the image to be recognized is extracted by an image outline feature grid, and a plurality of feature points on the outline feature curve are determined; Three feature points are selected from the plurality of feature points, and one feature point is selected from the three feature points as an intersection of two straight lines, a first straight line connected by the intersection and a first connecting point, and a second straight line connected by the intersection and a second connecting point are constructed, the first connecting point and the second connecting point are non-intersecting points in the three feature points, and the first connecting point is different from the second connecting point; When an included angle between the first straight line and the second straight line is greater than a preset included angle threshold, the outline feature curve is divided into a plurality of curve segments; A first fractal dimension and a second fractal dimension of each curve segment are calculated; When the first fractal dimensions of two adjacent curve segments are consistent, and the second fractal dimensions of the two curve segments are consistent, the two curve segments are merged to obtain the feedback image of the electric power tool; The calculation of the first fractal dimension and the second fractal dimension of each curve segment comprises the following steps: A first distance of each feature point on each curve segment to the first straight line is calculated; A second distance of each feature point on each curve segment to the second straight line is calculated; An average value of the first distances of the feature points on each curve segment is calculated, and the average value is determined as a first average error of the curve segment; An average value of the second distances of the feature points on each curve segment is calculated, and the average value is determined as a second average error of the curve segment; Based on the first average error of each curve segment and the number of curve segments of the outline feature curve, the first fractal dimension of the curve segment is calculated; Based on the second average error of each curve segment and the number of curve segments of the outline feature curve, the second fractal dimension of the curve segment is calculated.

2. The method of claim 1, wherein, Further comprising: When it is detected that the matching score of any electric power tool to be stored is less than the preset threshold score, a prompt information for reminding to supplement the electric power tool is sent to a management end.

3. The method of claim 1, wherein, The division of the outline feature curve into a plurality of curve segments comprises the following steps: The outline feature curve is divided into a plurality of curve segments with the feature points on the outline feature curve as the segmentation boundary points, and each curve segment contains at least two feature points.

4. The method of claim 1, wherein, Further comprising: When the included angle between the first straight line and the second straight line is not greater than the preset included angle threshold, a step of selecting three feature points from the plurality of feature points at random and selecting one feature point from the three feature points as an intersection point of two straight lines, constructing a first straight line obtained by connecting the intersection point and a first connection point, and a second straight line obtained by connecting the intersection point and a second connection point, is returned to be executed to update the first straight line and the second straight line, until the included angle between the updated first straight line and the updated second straight line is greater than the preset included angle threshold, the updated first straight line is determined as the latest first straight line, and the updated second straight line is determined as the latest second straight line.

5. The method according to any one of claims 1 to 4, characterized in that, For each product image of the to-be-warehoused power tool in the archive of the warehouse, a matching score of the feature matching between the feedback image and the product image is calculated, including: For each feature point of the contour feature curve in the feedback image, a target feature point matching the feature point in the product image of each to-be-warehoused power tool in the archive of the warehouse is determined; For each feature point of the contour feature curve in the feedback image, a matching sub-score between the feature point and the target feature point is calculated using the following formula, and is taken as the matching sub-score of the feature point: ； Where, a is the feature vector corresponding to a feature point of the contour feature curve in the feedback image, and b is the feature vector corresponding to the feature point in the product image that matches a, (x a y a (x) represents the feature point corresponding to the feature vector a. b y b ) represents the feature point corresponding to the feature vector b; The matching sub-scores of the feature points of the contour feature curve in the feedback image are accumulated to obtain a matching total score of the contour feature curve in the feedback image; The matching total score is divided by the number of feature points in the product image of each to-be-warehoused power tool to obtain the matching score of the feature matching between the feedback image and the product image of the to-be-warehoused power tool.

6. An electric power tool storage identification device characterized by comprising: The device is applied to the warehousing identification method of the power tool of claim 1, and the device comprises: An image acquisition unit is configured to acquire a to-be-identified image of the power tool when the power tool enters the warehouse; A feature point processing unit is configured to perform feature point processing on the to-be-identified image by a three-dimensional feature point algorithm to obtain a feedback image of the power tool; A matching score calculation unit is configured to calculate, for each product image of the to-be-warehoused power tool in the archive of the warehouse, a matching score of the feature matching between the feedback image and the product image, and take the matching score as the matching score of the to-be-warehoused power tool; A warehousing updating unit is configured to update the state of the to-be-warehoused power tool to warehoused when it is detected that the matching score of any to-be-warehoused power tool is not less than a preset threshold score.

7. A warehouse entry identification device for electrical tools, characterized in that, The device comprises a memory and a processor; The memory is configured to store a program; The processor is configured to execute the program to implement each step of the warehousing identification method of the power tool according to any one of claims 1-5.

8. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement each step of the warehousing identification method of the power tool according to any one of claims 1-5.

Images