CN111161346B - Method and device for layering commodities in goods shelves and electronic equipment - Google Patents
Method and device for layering commodities in goods shelves and electronic equipment Download PDFInfo
- Publication number
- CN111161346B CN111161346B CN201911398939.7A CN201911398939A CN111161346B CN 111161346 B CN111161346 B CN 111161346B CN 201911398939 A CN201911398939 A CN 201911398939A CN 111161346 B CN111161346 B CN 111161346B
- Authority
- CN
- China
- Prior art keywords
- commodity
- shelf
- shelf layer
- target
- goods
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
Abstract
The application discloses a method, a device and electronic equipment for layering commodities in a goods shelf. The method comprises the following steps: acquiring a shelf image; performing image visual analysis based on the shelf images, and determining the position information of each commodity in the shelf images; and judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods to the new goods shelf layer, otherwise, dividing the target goods to the matched goods shelf layer. The method has the advantages that the position information of the commodity can be automatically determined, the commodity is automatically distributed to the matched goods shelf layers according to the determined commodity position information, the influence of environmental factors such as light rays, angles and the like is not easy, the robustness is high, and the efficiency of goods shelf operation management can be greatly improved.
Description
Technical Field
The application relates to the field of image processing, in particular to a method, a device and electronic equipment for layering commodities in a goods shelf.
Background
In the conventional retail scenario, a lot of labor cost is required for operation, for example, the commodity is often placed in a shelf, and when a commodity on the shelf is sold, replenishment is required. In order to facilitate management and save labor cost, the computer vision technology is increasingly widely applied to retail scenes at present to improve operation management efficiency, for example, a shelf is shot, and the obtained shelf image is analyzed by using the computer vision technology to perform structured data expression on the shelf and commodities. However, the processing efficiency and accuracy of the structured data expression by using the prior art cannot meet the actual application requirements.
Disclosure of Invention
The present application has been made in view of the above problems, and provides a method, apparatus and electronic device for layering items in shelves that overcomes or at least partially solves the above problems.
According to one aspect of the present application there is provided a method of layering items in shelves comprising:
acquiring a shelf image;
performing image visual analysis based on the shelf images, and determining the position information of each commodity in the shelf images;
and judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods to the new goods shelf layer, otherwise, dividing the target goods to the matched goods shelf layer.
Optionally, the image visual analysis based on the shelf image, and determining the location information of each commodity in the shelf image includes:
detecting commodity bounding boxes corresponding to the commodities respectively from the shelf images by utilizing a commodity position detection model, and determining coordinates of vertexes of the commodity bounding boxes;
and calculating the coordinates of the center point of the commodity surrounding frame according to the coordinates of each vertex of the commodity surrounding frame, and taking the coordinates of the center point of the commodity surrounding frame as the position information of the corresponding commodity.
Optionally, the determining whether a new shelf layer needs to be built for the target commodity based on the position information of the target commodity includes:
sequencing all commodities according to the height position information in the position information to obtain a commodity sequence;
sequentially selecting target commodities from the commodity sequence, and determining the relative slope of the target commodity and each commodity in the current shelf layer according to the position information of the target commodity and the position information of each commodity in the current shelf layer;
and if the variance of each relative slope is not greater than the first threshold, taking the current shelf layer as the shelf layer matched with the target commodity.
Optionally, the determining whether a new shelf layer needs to be built for the target commodity based on the location information of the target commodity further includes:
if the variance of each relative slope is larger than a first threshold and the relative slope with the largest absolute value is larger than a second threshold, judging whether the height difference between the commodity in the current goods shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a third threshold; the third threshold is determined according to the minimum height value of each commodity detected from the shelf image;
if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
Optionally, the determining whether a new shelf layer needs to be built for the target commodity based on the location information of the target commodity further includes:
if the variance of each relative slope is larger than the first threshold and the relative slope with the largest absolute value is not larger than the second threshold, judging whether the relative slope with the largest absolute value is larger than the fourth threshold;
and if the current goods shelf layer is not larger than the target goods, taking the current goods shelf layer as the goods shelf layer matched with the target goods.
Optionally, the determining whether a new shelf layer needs to be built for the target commodity based on the location information of the target commodity further includes:
if the relative slope with the largest absolute value is larger than a fourth threshold, judging whether the transverse difference between the commodity in the current goods shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a fifth threshold; the fifth threshold is determined based on a minimum width of each commodity detected from the shelf image;
if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
Optionally, the method further comprises:
if the number of the commodities in the current shelf layer reaches a sixth threshold value, calculating the slope of the current shelf layer according to the position information of the commodities in the current shelf layer;
If the slope direction of the current shelf layer is the same as the slope direction of the adjacent shelf layer of the current shelf layer, and the slope of the current shelf layer is not smaller than the slope of the adjacent shelf layer of the current shelf layer, updating the position information of each commodity divided to the current shelf layer according to the slope of the adjacent shelf layer of the current shelf layer.
According to another aspect of the present application there is provided an apparatus for layering items of merchandise in a shelf, comprising:
the acquisition unit is used for acquiring the shelf images;
the image visual analysis unit is used for carrying out image visual analysis based on the shelf images and determining the position information of each commodity in the shelf images;
and the layering unit is used for judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods into the new goods shelf layer, and if not, dividing the target goods into the matched goods shelf layers.
Optionally, the visual analysis unit is configured to detect a commodity bounding box corresponding to each commodity from the shelf image by using a commodity position detection model, and determine coordinates of each vertex of the commodity bounding box; and calculating the coordinates of the center point of the commodity surrounding frame according to the coordinates of each vertex of the commodity surrounding frame, and taking the coordinates of the center point of the commodity surrounding frame as the position information of the corresponding commodity.
Optionally, the layering unit is used for ordering the commodities according to the height position information in the position information to obtain a commodity sequence; sequentially selecting target commodities from the commodity sequence, and determining the relative slope of the target commodity and each commodity in the current shelf layer according to the position information of the target commodity and the position information of each commodity in the current shelf layer; and if the variance of each relative slope is not greater than the first threshold, taking the current shelf layer as the shelf layer matched with the target commodity.
Optionally, the layering unit is configured to determine whether a height difference between the commodity in the current shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is greater than a third threshold if the variance of each relative slope is greater than a first threshold and the relative slope with the largest absolute value is greater than a second threshold; the third threshold is determined according to the minimum height value of each commodity detected from the shelf image; if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
Optionally, the layering unit is configured to determine whether the relative slope with the largest absolute value is greater than the fourth threshold if the variance of each relative slope is greater than the first threshold and the relative slope with the largest absolute value is not greater than the second threshold; and if the current goods shelf layer is not larger than the target goods, taking the current goods shelf layer as the goods shelf layer matched with the target goods.
Optionally, the layering unit is configured to determine whether a lateral difference between the commodity in the current shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is greater than a fifth threshold if the relative slope with the largest absolute value is greater than the fourth threshold; the fifth threshold is determined based on a minimum width of each commodity detected from the shelf image; if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
Optionally, the apparatus further includes: the updating unit is used for calculating the slope of the current goods shelf layer according to the position information of the goods in the current goods shelf layer if the number of the goods in the current goods shelf layer reaches a sixth threshold value; if the slope direction of the current shelf layer is the same as the slope direction of the adjacent shelf layer of the current shelf layer, and the slope of the current shelf layer is not smaller than the slope of the adjacent shelf layer of the current shelf layer, updating the position information of each commodity divided to the current shelf layer according to the slope of the adjacent shelf layer of the current shelf layer.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory arranged to store computer executable instructions that, when executed, cause the processor to perform a method as described in any of the above.
According to a further aspect of the present application there is provided a computer readable storage medium storing one or more programs which when executed by a processor implement a method as described in any of the above.
From the above, the technical solution of the present application is to obtain the shelf image; performing image visual analysis based on the shelf images, and determining the position information of each commodity in the shelf images; and judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods to the new goods shelf layer, otherwise, dividing the target goods to the matched goods shelf layer. The method has the advantages that the position information of the commodity can be automatically determined, the commodity is automatically distributed to the matched goods shelf layers according to the determined commodity position information, the influence of environmental factors such as light rays, angles and the like is not easy, the robustness is high, and the efficiency of goods shelf operation management can be greatly improved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 illustrates a flow diagram of a method of layering items in shelves according to one embodiment of the application;
FIG. 2 shows a schematic of an apparatus for layering items in shelves according to one embodiment of the application;
FIG. 3 illustrates a schematic diagram of processing logic for layering items in shelves according to one embodiment of the application;
FIG. 4 shows a schematic structural diagram of an electronic device according to one embodiment of the application;
FIG. 5 illustrates a schematic diagram of a structure of a computer-readable storage medium according to one embodiment of the present application;
FIG. 6 shows a shelf image taken of a shelf at a large angle;
fig. 7 shows another shelf image taken of a shelf at a large angle.
Detailed Description
Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.
In order to realize data structuring of commodities and obtain layering information of the commodities in a goods shelf, the goods shelf can be layered by using a method of label backboard information, the position of the label backboard and basic image characteristics such as brightness, color, direction and gradient on a goods shelf image are detected through extracting a label template, and then the position information of the label backboard is obtained through comparison with a preset coefficient threshold according to a correlation coefficient.
However, there are a plurality of drawbacks in this solution, for example, this solution relies on the detection effect of the label back plate, but in the practical application scenario, the situation that there is no label back plate in the shelf is also common; even if a tag backboard exists, the shooting is easily influenced by environmental factors such as light rays, angles and the like, so that the robustness is poor; the goods shelf layering cannot be carried out by directly utilizing the label backboard information, so that layering information of goods in the goods shelf is determined; there is a large limitation on the shooting angle, for example, there is a good effect on the shelf image shot in front without deflection of the tilting angle, but in the case of shooting at a large angle, it is difficult to simply judge the shelf image according to the scheme.
The application provides a scheme for layering commodities in a goods shelf based on deep learning, which can be widely applied to scenes, is more robust, and can be applied to various complex scenes such as large-angle shooting, multi-row commodities on the same layer and the like. The technical scheme of the application is described in detail below by way of examples.
FIG. 1 shows a flow diagram of a method of layering items in shelves, according to one embodiment of the application, as shown in FIG. 1, comprising:
step S110, acquiring a shelf image.
The goods shelf image can reflect the specific information such as the quantity information of goods, the position information of goods, the goods shelf layer and the like, and a foundation can be provided for subsequent data analysis by acquiring the goods shelf image.
Step S120, performing image visual analysis based on the shelf images, and determining the position information of each commodity in the shelf images.
The image visual analysis is to process an image by CV (Computational Vision, computer vision), for example, to perform semantic recognition, to determine the type and number of articles included in the image, and to perform labeling.
In this step, the position information of each commodity in the shelf image may be detected according to a shelf commodity position detection model, which may be obtained by training based on commodity position labeling data, to obtain commodity position frame bbox information in the image, and specifically expressed as [ x ] in terms of coordinates 0 ,y 0 ,x 1 ,y 1 ]Wherein (x) 0 ,y 0 ) Is the coordinate of the upper left corner point of the commodity frame, (x) 1 ,y 1 ) Is the coordinates of the right lower corner of the commodity frame. In this way, the position information of the commodity can be determined through the commodity position detection model of the goods shelf.
Step S130, judging whether a new shelf layer needs to be established for the target commodity based on the position information of the target commodity, if so, dividing the target commodity into the new shelf layer, and if not, dividing the target commodity into the matched shelf layer.
It may be determined whether a new shelf layer needs to be established for the target commodity based on the location information of the target commodity. For example, the positions of the commodity center point and all commodity center points in the previous layer can be compared with a preset threshold value based on the slope, the height difference and the level difference, so that whether a new layer is established or not can be judged, when the judging result is that a new goods shelf layer needs to be established for the target commodity, the target commodity can be divided into the new goods shelf layer, and when the judging result is that the new goods shelf layer does not need to be established for the target commodity, the target commodity can be divided into the matched goods shelf layers. In this way, according to the correlation coefficient, whether a new shelf layer needs to be established for the target commodity can be judged by comparing the correlation coefficient with a preset coefficient threshold value.
Therefore, the method shown in fig. 1 can automatically determine the position information of the commodity, automatically divide the commodity to the matched shelf layers according to the determined commodity position information, is not easy to be influenced by environmental factors such as light rays, angles and the like, has strong robustness, and can greatly improve the efficiency of shelf operation management.
In one embodiment of the present application, in the method, performing image visual analysis based on the shelf image, determining the location information of each commodity in the shelf image includes: detecting commodity bounding boxes corresponding to the commodities respectively from the shelf images by utilizing a commodity position detection model, and determining coordinates of vertexes of the commodity bounding boxes; and calculating the coordinates of the center point of the commodity bounding box according to the coordinates of each vertex of the commodity bounding box, and taking the coordinates of the center point of the commodity bounding box as the position information of the corresponding commodity.
In order to accurately acquire the position information of the commodity, the commodity position detection model can be used for detecting commodity bounding boxes corresponding to the commodities respectively from the shelf image, and determining the coordinates of the vertexes of the commodity bounding boxes, for example, the commodity position box bbox information, [ x ] in the image can be acquired 0 ,y 0 ,x 1 ,y 1 ]Wherein (x) 0 ,y 0 ) Is the coordinate of the upper left corner point of the commodity frame, (x) 1 ,y 1 ) Is the coordinates of the right lower corner of the commodity frame. The coordinates of the center point of the commodity enclosure frame may be calculated according to the coordinates of the vertices of the commodity enclosure frame, and the coordinates of the center point of the commodity enclosure frame may be used as the position information of the corresponding commodity. For example, the commodity center point position information [ x, y ] may be obtained from the commodity position frame bbox information ]Wherein x= (x 0 +x 1 )/2，y＝(y 0 +y 1 ) And/2, thus taking the central point position as the layered basic information can make the application effect more robust.
In an embodiment of the present application, in the above method, determining whether a new shelf layer needs to be built for the target commodity based on the position information of the target commodity includes: sequencing the commodities according to the height position information in the position information to obtain a commodity sequence; sequentially selecting target commodities from the commodity sequence, and determining the relative slope of the target commodity and each commodity in the current shelf layer according to the position information of the target commodity and the position information of each commodity in the current shelf layer; and if the variance of each relative slope is not greater than the first threshold, taking the current shelf layer as the shelf layer matched with the target commodity.
Under the condition that a certain inclination angle exists during shooting, due to the close-large-far-small space proportion relation, various commodities in the image can be at different heights. Based on the height position information of the commodities, the commodities can be ordered, and then the goods shelf layer where each commodity is located is judged.
Specifically, when the first commodity (i.e., the commodity with the highest relative position in the image) is selected, since there is no shelf layer yet, a new one is created as the current shelf layer, and the current shelf layer is the shelf layer that matches the first commodity.
For the second commodity, the relative slope between the second commodity and the first commodity is calculated, and since only one relative slope is at this time, the variance is 0 and is not greater than the first threshold, the current shelf layer is also the shelf layer matched with the second commodity.
For the third commodity, calculating the relative slope between the third commodity and the first commodity and the second commodity respectively, and calculating the variance of the relative slope. If the variance of each relative slope is not greater than a first threshold (e.g., 0.1), then the current shelf level is also the shelf level that matches the third item. The above procedure may also be repeated for subsequent goods.
If the variance of each relative slope is greater than the first threshold, a new shelf layer can be directly established for the target commodity, and further judgment can be performed. In an embodiment of the present application, in the above method, determining whether a new shelf layer needs to be built for the target commodity based on the position information of the target commodity further includes: if the variance of each relative slope is larger than a first threshold and the relative slope with the largest absolute value is larger than a second threshold, judging whether the height difference between the commodity in the current shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a third threshold; the third threshold is determined based on the minimum height of each commodity detected from the shelf image; if so, a new shelf layer is established for the target commodity, otherwise, the current shelf layer is used as the shelf layer matched with the target commodity.
As shown in fig. 6, each thin frame is a commodity position frame, and is labeled with one commodity, it can be seen that, due to the large-angle shooting, not only the commodity in the front row but also the commodity in the rear row in the same shelf layer are shot in the thick frame, and the commodity in the rear row is also labeled. However, due to the reasons of shielding, perspective and the like, the commodity number of the rear-row commodity is the same as that of the front-row commodity (the commodity numbers are the same), but the sizes of the commodity position frames are different, and particularly, the height positions of the center points are different to a certain extent.
From another perspective, the height position difference of the front-back row of commodities on the same layer is smaller than the height position difference of the commodities on different layers, and the goods shelf layers can be accurately divided according to the characteristic.
That is, although the calculated variance of each relative slope may be greater than the first threshold for the front and rear row products of the same tier, it may be determined whether the absolute value of the largest relative slope is greater than the second threshold (e.g., set to 1), as can be seen in fig. 6, where one rear row product is located in the same row as its adjacent front row product, but the slope is obviously great. If this is the case, it is further determined whether the difference in height between the commodity in the current shelf layer corresponding to the maximum relative slope of the absolute value and the target commodity is greater than a third threshold, where the third threshold may be determined according to the minimum height value of each commodity detected from the shelf image, for example, 0.5 x the goods_min_height, where the goods_min_height is the minimum height value of each commodity detected from the shelf image. If the height difference between the commodity in the current commodity shelf layer and the target commodity corresponding to the maximum relative slope of the absolute value is not greater than the third threshold value, the height difference is not yet up to the degree of difference between the two commodity shelf layers (the height difference between the two commodity shelf layers can be obtained by making a difference between the center point of the upper commodity and the center point of the lower commodity and is at least greater than the self height of half commodity), that is, the height difference is possibly caused by shooting the commodity in the rear row.
In an embodiment of the present application, in the above method, determining whether a new shelf layer needs to be built for the target commodity based on the position information of the target commodity further includes: if the variance of each relative slope is larger than the first threshold and the relative slope with the largest absolute value is not larger than the second threshold, judging whether the relative slope with the largest absolute value is larger than the fourth threshold; and if the current goods shelf layer is not larger than the target goods, taking the current goods shelf layer as the goods shelf layer matched with the target goods.
For example, if the variance grad_std of each relative slope is greater than a first threshold (e.g., 0.1) and the absolute value of the maximum relative slope max_abs_grad is not greater than a fourth threshold (e.g., 0.5), the current shelf layer is the shelf layer matching the target commodity. In this case, although the variance of the relative slopes is large, the overall slope of the shelf layer itself may be large due to the large angle shooting, and thus there may be a certain difference in the height position difference between the products in the shelf layer. At this time, if the relative slope with the largest absolute value is not very large, it is indicated that the slope between the shelf layers has not been reached yet, and thus the current shelf layer is divided.
In the other case, as shown in fig. 7, it can be seen that the slope of the center point of the commodity at both ends of the same shelf may be relatively large due to the large angle of the shelf layer shown by the arrow in fig. 7. However, it can be seen that the situation is obviously different from the situation of the goods shelves on different layers, namely, the level difference of the central positions of goods on different layers of goods shelves is small, and the goods shelves can be accurately divided according to the characteristic.
In an embodiment of the present application, in the above method, determining whether a new shelf layer needs to be built for the target commodity based on the position information of the target commodity further includes: if the relative slope with the largest absolute value is larger than a fourth threshold value, judging whether the transverse difference between the commodity in the current shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a fifth threshold value; the fifth threshold is determined based on the minimum width of each commodity detected from the shelf image; if so, a new shelf layer is established for the target commodity, otherwise, the current shelf layer is used as the shelf layer matched with the target commodity.
For example, if the overall slope of a shelf is already greater than the fourth threshold (e.g., 0.5) for imaging reasons, it is also possible that the absolute maximum relative slope calculated for a target commodity in the shelf is greater than the fourth threshold. At this time, the judgment can be performed according to the characteristic that the level difference of the commodity center positions of different shelf layers is smaller, namely, whether the transverse difference between the commodity in the current shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a fifth threshold value is judged. If the gap is larger, the commodity is indicated to be the same layer; if the gap is small, it indicates that the commodities are commodities in different layers (for example, the commodities in the lower layer adjacent to the commodities in the upper layer satisfy the characteristics of large relative slope but small level difference).
Fig. 3 is a schematic diagram of processing logic for layering commodities in a shelf according to an embodiment of the present application, where the processing logic integrates the above embodiments, and as a preferred processing manner, as shown in fig. 3, grad_list is a set of relative slopes between a newly added commodity center point and all commodity center points of a current shelf layer, grad_std is a variance of grad_list, max_abs_grad is a maximum value of absolute values in grad_list, goods_min_height is a minimum value of heights of all commodities detected in an image, goods_min_width is a minimum value of widths of all commodities detected in an image, max_grad_height is a height difference (y coordinate difference) of a commodity corresponding to max_abs_grad of a target commodity, and max_grad_width is a lateral difference (x coordinate difference) of a commodity corresponding to max_abs_grad.
According to fig. 3, when the first commodity (i.e., the commodity with the highest relative position in the image) is selected, since there is no shelf layer yet, a new one is created as the current shelf layer, and the current shelf layer is the shelf layer that matches the first commodity.
For the second commodity, the relative slope between the second commodity and the first commodity is calculated, and since only one relative slope is at this time, the variance is 0 and is not more than 0.1, the current shelf layer is also the shelf layer matched with the second commodity.
For the third commodity, calculating the relative slope grad_list between the third commodity and the first commodity and the second commodity, and calculating grad_std.
If grad_std is not greater than 0.1, then the current shelf level is also the shelf level that matches the third item.
If the grad_std is greater than 0.1, judging whether the max_abs_grad is greater than 1, if so, describing that the situation of fig. 6 may occur, judging whether the max_grad_height is greater than 0.5 x good_min_height, if so, creating a shelf layer, otherwise, the current shelf layer is also a shelf layer matched with the third commodity.
If max_abs_grad is not greater than 1, then the situation of FIG. 7 may occur, determining if max_abs_grad is greater than 0.5, and if not, the current shelf layer is also the shelf layer that matches the third item. If so, judging whether the max_grad_width is larger than 2 x good_min_width, if so, newly building a shelf layer, otherwise, the current shelf layer is a shelf layer matched with the third commodity.
The following commodities can be similarly calculated, and are not described herein.
In one embodiment of the present application, the method further comprises: if the number of the commodities in the current shelf layer reaches a sixth threshold value, calculating the slope of the current shelf layer according to the position information of the commodities in the current shelf layer; if the slope direction of the current shelf layer is the same as the slope direction of the adjacent shelf layer of the current shelf layer, and the slope of the current shelf layer is not smaller than the slope of the adjacent shelf layer of the current shelf layer, updating the position information of each commodity distributed to the current shelf layer according to the slope of the adjacent shelf layer of the current shelf layer.
To obtain more unified and concise data, the center point of the commodity can be corrected. For example, a sixth threshold value reached by the number of products in the current shelf layer may be preset, and the slope of the current shelf layer may be calculated according to the position information of the products in the current shelf layer. Then, whether to correct the center point of the commodity position is judged according to the relation between the slope magnitude and the direction.
For example, the slope next_delta_y of the commodity center point of each layer of the commodity shelf and the slope delta_y of the commodity center point of the previous layer of the commodity shelf can be calculated based on the first two commodity position center points of each layer of the commodity shelf, and if any one of the following conditions is met, no correction is performed: (1) next_delta_y <0, i.e. the slope directions of adjacent two layers of goods shelf goods are opposite; (2) abs (next_delta_y) < abs (delta_y), i.e., the slope of the upper pallet is smaller than that of the upper pallet. Otherwise, the center point position is corrected.
After the judgment result to be corrected is obtained, the positions of all the remaining commodity center points can be corrected based on the slope delta_y of the upper layer shelf, and the following formula is adopted: new_y=y-delta_y (x-x 0). Wherein x0 is the abscissa of the leftmost commodity center point of the previous row, [ x, y ] is the commodity center point position to be corrected, and the corrected commodity center point position is [ x, new_y ]. In this way, whether to perform position correction on the remaining commodity center points which are not subjected to layer division can be determined based on the direction and magnitude relation of the slopes of the commodity of the new layer and the commodity of the previous layer, and the position correction is performed based on a slope formula.
FIG. 2 shows a schematic structural diagram of an apparatus for layering items in shelves according to one embodiment of the present application, and as shown in FIG. 2, an apparatus 200 for layering items in shelves includes:
and an acquisition unit 210, configured to acquire a shelf image.
The goods shelf image can reflect the specific information such as the quantity information of goods, the position information of goods, the goods shelf layer and the like, and a foundation can be provided for subsequent data analysis by acquiring the goods shelf image.
The image visual analysis unit 220 is configured to perform image visual analysis based on the shelf image, and determine positional information of each commodity in the shelf image.
The image visual analysis is to process an image by CV (Computational Vision, computer vision), for example, to perform semantic recognition, to determine the type and number of articles included in the image, and to perform labeling.
The image visual analysis unit 220 may specifically detect the position information of each commodity in the shelf image according to a shelf commodity position detection model, where the commodity position detection model may be obtained by training based on commodity position labeling data, and obtain commodity position frame bbox information in the image, and specifically expressed as [ x ] in terms of coordinates 0 ,y 0 ,x 1 ,y 1 ]Wherein (x) 0 ,y 0 ) Is the coordinate of the upper left corner point of the commodity frame, (x) 1 ,y 1 ) Is the coordinates of the right lower corner of the commodity frame. In this way, the position information of the commodity can be determined through the commodity position detection model of the goods shelf.
And the layering unit 230 is configured to determine whether a new shelf layer needs to be established for the target commodity based on the position information of the target commodity, if so, divide the target commodity into the new shelf layer, and if not, divide the target commodity into the matched shelf layer.
It may be determined whether a new shelf layer needs to be established for the target commodity based on the location information of the target commodity. For example, the positions of the commodity center point and all commodity center points in the previous layer can be compared with a preset threshold value based on the slope, the height difference and the level difference, so that whether a new layer is established or not can be judged, when the judging result is that a new goods shelf layer needs to be established for the target commodity, the target commodity can be divided into the new goods shelf layer, and when the judging result is that the new goods shelf layer does not need to be established for the target commodity, the target commodity can be divided into the matched goods shelf layers. In this way, according to the correlation coefficient, whether a new shelf layer needs to be established for the target commodity can be judged by comparing the correlation coefficient with a preset coefficient threshold value.
Therefore, the device shown in fig. 2 can automatically determine the position information of the commodity, automatically divide the commodity to the matched shelf layers according to the determined commodity position information, is not easy to be influenced by environmental factors such as light rays, angles and the like, has strong robustness, and can greatly improve the efficiency of shelf operation management.
In one embodiment of the present application, in the above-described apparatus, the image visual analysis unit 220 is configured to detect, from the shelf image, a commodity bounding box corresponding to each commodity, respectively, using the commodity position detection model, and determine coordinates of each vertex of the commodity bounding box; and calculating the coordinates of the center point of the commodity bounding box according to the coordinates of each vertex of the commodity bounding box, and taking the coordinates of the center point of the commodity bounding box as the position information of the corresponding commodity.
In an embodiment of the present application, in the above apparatus, the layering unit 230 is configured to sort each commodity according to height position information in the position information, so as to obtain a commodity sequence; sequentially selecting target commodities from the commodity sequence, and determining the relative slope of the target commodity and each commodity in the current shelf layer according to the position information of the target commodity and the position information of each commodity in the current shelf layer; and if the relative slope with the maximum absolute value is not greater than the first threshold value, taking the current shelf layer as the shelf layer matched with the target commodity.
In an embodiment of the present application, in the above apparatus, the layering unit 230 is configured to determine whether a height difference between a commodity in a current shelf layer corresponding to a maximum absolute value relative slope and a target commodity is greater than a third threshold if the maximum absolute value relative slope is greater than a first threshold and the maximum absolute value relative slope is greater than a second threshold; the third threshold is determined based on the minimum height of each commodity detected from the shelf image; if so, a new shelf layer is established for the target commodity, otherwise, the current shelf layer is used as the shelf layer matched with the target commodity.
In an embodiment of the present application, in the foregoing apparatus, the layering unit 230 is configured to determine whether the absolute value of the relative slope is greater than the fourth threshold if the absolute value of the relative slope is greater than the first threshold and the absolute value of the relative slope is not greater than the second threshold; and if the current goods shelf layer is not larger than the target goods, taking the current goods shelf layer as the goods shelf layer matched with the target goods.
In an embodiment of the present application, in the above apparatus, the layering unit 230 is configured to determine whether a lateral difference between the commodity in the current shelf layer corresponding to the maximum absolute value relative slope and the target commodity is greater than a fifth threshold if the maximum absolute value relative slope is greater than the fourth threshold; the fifth threshold is determined based on the minimum width of each commodity detected from the shelf image; if so, a new shelf layer is established for the target commodity, otherwise, the current shelf layer is used as the shelf layer matched with the target commodity.
In one embodiment of the present application, the apparatus further comprises: the updating unit is used for calculating the slope of the current goods shelf layer according to the position information of the goods in the current goods shelf layer if the number of the goods in the current goods shelf layer reaches a sixth threshold value; if the slope direction of the current shelf layer is the same as the slope direction of the adjacent shelf layer of the current shelf layer, and the slope of the current shelf layer is not smaller than the slope of the adjacent shelf layer of the current shelf layer, updating the position information of each commodity distributed to the current shelf layer according to the slope of the adjacent shelf layer of the current shelf layer.
It should be noted that, the specific implementation manner of each embodiment of the apparatus may be performed with reference to the specific implementation manner of the corresponding embodiment of the method, which is not described herein.
In summary, according to the technical scheme of the application, the shelf image is obtained; performing image visual analysis based on a shelf image, and determining the position information of each commodity in the shelf image; and judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods to the new goods shelf layer, otherwise, dividing the target goods to the matched goods shelf layer. The method has the advantages that the position information of the commodity can be automatically determined, the commodity is automatically distributed to the matched goods shelf layers according to the determined commodity position information, the influence of environmental factors such as light rays, angles and the like is not easy, the robustness is high, and the efficiency of goods shelf operation management can be greatly improved.
It should be noted that:
the algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose devices may also be used with the teachings herein. The required structure for the construction of such devices is apparent from the description above. In addition, the present application is not directed to any particular programming language. It will be appreciated that the teachings of the present application described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present application.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the above description of exemplary embodiments of the application, various features of the application are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed application requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this application.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the application and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
Various component embodiments of the application may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functions of some or all of the components in a shelf layering device according to embodiments of the present application may be implemented in practice using a microprocessor or Digital Signal Processor (DSP). The present application can also be implemented as an apparatus or device program (e.g., a computer program and a computer program product) for performing a portion or all of the methods described herein. Such a program embodying the present application may be stored on a computer readable medium, or may have the form of one or more signals. Such signals may be downloaded from an internet website, provided on a carrier signal, or provided in any other form.
For example, fig. 4 shows a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device 400 comprises a processor 410 and a memory 420 arranged to store computer executable instructions (computer readable program code). The memory 420 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. The memory 420 has a storage space 430 storing computer readable program code 431 for performing any of the method steps described above. For example, the memory space 430 for storing computer readable program code may include individual computer readable program code 431 for implementing the various steps in the above methods, respectively. The computer readable program code 431 may be read from or written to one or more computer program products. These computer program products comprise a program code carrier such as a hard disk, a Compact Disc (CD), a memory card or a floppy disk. Such a computer program product is typically a computer readable storage medium as described for example in fig. 5. Fig. 5 illustrates a schematic structure of a computer-readable storage medium according to an embodiment of the present application. The computer readable storage medium 500 stores computer readable program code 431 for performing the steps of the method according to the present application, which may be read by the processor 410 of the electronic device 400, which computer readable program code 431, when executed by the electronic device 400, causes the electronic device 400 to perform the steps of the method described above, in particular, the computer readable program code 431 stored by the computer readable storage medium may perform the method shown in any of the embodiments described above. The computer readable program code 431 may be compressed in a suitable form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The application may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names.
Claims (7)
1. A method of layering items in a shelf, comprising: acquiring a shelf image; performing image visual analysis based on the shelf images, and determining the position information of each commodity in the shelf images; judging whether a new goods shelf layer is required to be established for the target goods based on the position information of the target goods, if so, dividing the target goods into the new goods shelf layer, otherwise, dividing the target goods into the matched goods shelf layers;
Wherein, based on the position information of the target commodity, determining whether a new shelf layer needs to be established for the target commodity includes: sequencing all commodities according to the height position information in the position information to obtain a commodity sequence; sequentially selecting target commodities from the commodity sequence, and determining the relative slope of the target commodity and each commodity in the current shelf layer according to the position information of the target commodity and the position information of each commodity in the current shelf layer; if the variance of each relative slope is not greater than a first threshold, taking the current shelf layer as the shelf layer matched with the target commodity;
the determining whether a new shelf layer needs to be established for the target commodity based on the position information of the target commodity further comprises: if the variance of each relative slope is larger than a first threshold and the relative slope with the largest absolute value is larger than a second threshold, judging whether the height difference between the commodity in the current goods shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a third threshold; the third threshold is determined according to the minimum height value of each commodity detected from the shelf image; if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
2. The method of claim 1, wherein the determining location information for each item in the shelf image based on the visual analysis of the image of the shelf comprises: detecting commodity bounding boxes corresponding to the commodities respectively from the shelf images by utilizing a commodity position detection model, and determining coordinates of vertexes of the commodity bounding boxes; and calculating the coordinates of the center point of the commodity surrounding frame according to the coordinates of each vertex of the commodity surrounding frame, and taking the coordinates of the center point of the commodity surrounding frame as the position information of the corresponding commodity.
3. The method of claim 2, wherein determining whether a new shelf layer needs to be established for the target commodity based on the location information of the target commodity further comprises: if the variance of each relative slope is larger than the first threshold and the relative slope with the largest absolute value is not larger than the second threshold, judging whether the relative slope with the largest absolute value is larger than the fourth threshold; and if the current goods shelf layer is not larger than the target goods, taking the current goods shelf layer as the goods shelf layer matched with the target goods.
4. The method of claim 3, wherein determining whether a new shelf layer needs to be established for the target commodity based on the location information of the target commodity further comprises: if the relative slope with the largest absolute value is larger than a fourth threshold, judging whether the transverse difference between the commodity in the current goods shelf layer corresponding to the relative slope with the largest absolute value and the target commodity is larger than a fifth threshold; the fifth threshold is determined based on a minimum width of each commodity detected from the shelf image; if so, establishing a new shelf layer for the target commodity, otherwise, taking the current shelf layer as a shelf layer matched with the target commodity.
5. The method of any one of claims 1-4, wherein the method further comprises: if the number of the commodities in the current shelf layer reaches a sixth threshold value, calculating the slope of the current shelf layer according to the position information of the commodities in the current shelf layer; if the slope direction of the current shelf layer is the same as the slope direction of the adjacent shelf layer of the current shelf layer, and the slope of the current shelf layer is not smaller than the slope of the adjacent shelf layer of the current shelf layer, updating the position information of each commodity divided to the current shelf layer according to the slope of the adjacent shelf layer of the current shelf layer.
6. An electronic device, wherein the electronic device comprises: a processor; and a memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method of any of claims 1-5.
7. A computer readable storage medium storing one or more programs which, when executed by a processor, implement the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911398939.7A CN111161346B (en) | 2019-12-30 | 2019-12-30 | Method and device for layering commodities in goods shelves and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911398939.7A CN111161346B (en) | 2019-12-30 | 2019-12-30 | Method and device for layering commodities in goods shelves and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111161346A CN111161346A (en) | 2020-05-15 |
| CN111161346B true CN111161346B (en) | 2023-09-12 |
Family
ID=70559606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911398939.7A Active CN111161346B (en) | 2019-12-30 | 2019-12-30 | Method and device for layering commodities in goods shelves and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111161346B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113191210B (en) * | 2021-04-09 | 2023-08-29 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment |
| CN113140073B (en) * | 2021-05-11 | 2022-12-13 | 支付宝(杭州)信息技术有限公司 | Layer control method and system of intelligent container, and intelligent container |
| CN113569830B (en) * | 2021-06-22 | 2024-08-02 | 邬国锐 | Method, apparatus, device and storage medium for determining display article row and column position |
| CN113610462A (en) * | 2021-07-29 | 2021-11-05 | 华清科盛(北京)信息技术有限公司 | Error searching method and system for lightweight warehouse logistics center, intelligent terminal and computer readable storage medium |
| CN114140676A (en) * | 2021-11-03 | 2022-03-04 | 上海小零网络科技有限公司 | Method, device and medium for determining shelf layer based on image recognition |
| CN114529841B (en) * | 2022-02-23 | 2025-08-19 | 上海汉时信息科技有限公司 | Shelf layering method and device |
| CN116129336A (en) * | 2023-01-03 | 2023-05-16 | 北京朗镜科技有限责任公司 | Method, device, equipment and storage medium for detecting goods on goods shelf |
| CN115983951B (en) * | 2023-03-20 | 2023-12-19 | 北京鲜衣怒马文化传媒有限公司 | Virtual space shelf commodity arrangement method, system and readable storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2014192612A1 (en) * | 2013-05-31 | 2017-02-23 | 日本電気株式会社 | Image recognition apparatus, processing method thereof, and program |
| CN106781014A (en) * | 2017-01-24 | 2017-05-31 | 广州市蚁道互联网有限公司 | Automatic vending machine and its operation method |
| CN107292248A (en) * | 2017-06-05 | 2017-10-24 | 广州诚予国际市场信息研究有限公司 | A kind of merchandise control method and system based on image recognition technology |
| CN107463945A (en) * | 2017-07-12 | 2017-12-12 | 浙江大学 | A kind of type of merchandize recognition methods based on depth matching network |
| CN107833363A (en) * | 2017-09-28 | 2018-03-23 | 中南大学 | Self-help vending machine automated stock control system and method based on image recognition |
| CN108764778A (en) * | 2018-04-26 | 2018-11-06 | 东华大学 | A kind of warehouse unmanned plane supplement cargo method based on image procossing |
| CN109509304A (en) * | 2017-09-14 | 2019-03-22 | 阿里巴巴集团控股有限公司 | Automatic vending machine and its control method, device and computer system |
| CN109685978A (en) * | 2017-10-18 | 2019-04-26 | 杭州海康威视数字技术股份有限公司 | A kind of automatic goods loading system and automatic vending machine |
| CN109816862A (en) * | 2017-11-20 | 2019-05-28 | 上海孚迅机器人有限公司 | Can automatic recognition cargo position automatic vending machine |
| CN110033452A (en) * | 2019-04-17 | 2019-07-19 | 北京云量互动科技有限公司 | Refrigerator shelf commodity statistics method and system based on image processing techniques |
| CN110245580A (en) * | 2019-05-24 | 2019-09-17 | 北京百度网讯科技有限公司 | A method, apparatus, device and computer storage medium for detecting images |
| CN110472486A (en) * | 2019-07-03 | 2019-11-19 | 北京三快在线科技有限公司 | A kind of shelf obstacle recognition method, device, equipment and readable storage medium storing program for executing |
| CN110610149A (en) * | 2019-09-03 | 2019-12-24 | 卓尔智联(武汉)研究院有限公司 | An information processing method, device and computer storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8706293B2 (en) * | 2011-11-29 | 2014-04-22 | Cereson Co., Ltd. | Vending machine with automated detection of product position |
| US20160260049A1 (en) * | 2015-03-06 | 2016-09-08 | Wal-Mart Stores, Inc. | Shopping facility assistance systems, devices, and methods to dispatch and recover motorized transport units that effect remote deliveries |
| US9928438B2 (en) * | 2016-03-10 | 2018-03-27 | Conduent Business Services, Llc | High accuracy localization system and method for retail store profiling via product image recognition and its corresponding dimension database |
| US10198653B2 (en) * | 2017-04-26 | 2019-02-05 | Sensors Incorporated | System and method for performing production line product identification |
-
2019
- 2019-12-30 CN CN201911398939.7A patent/CN111161346B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPWO2014192612A1 (en) * | 2013-05-31 | 2017-02-23 | 日本電気株式会社 | Image recognition apparatus, processing method thereof, and program |
| CN106781014A (en) * | 2017-01-24 | 2017-05-31 | 广州市蚁道互联网有限公司 | Automatic vending machine and its operation method |
| CN107292248A (en) * | 2017-06-05 | 2017-10-24 | 广州诚予国际市场信息研究有限公司 | A kind of merchandise control method and system based on image recognition technology |
| CN107463945A (en) * | 2017-07-12 | 2017-12-12 | 浙江大学 | A kind of type of merchandize recognition methods based on depth matching network |
| CN109509304A (en) * | 2017-09-14 | 2019-03-22 | 阿里巴巴集团控股有限公司 | Automatic vending machine and its control method, device and computer system |
| CN107833363A (en) * | 2017-09-28 | 2018-03-23 | 中南大学 | Self-help vending machine automated stock control system and method based on image recognition |
| CN109685978A (en) * | 2017-10-18 | 2019-04-26 | 杭州海康威视数字技术股份有限公司 | A kind of automatic goods loading system and automatic vending machine |
| CN109816862A (en) * | 2017-11-20 | 2019-05-28 | 上海孚迅机器人有限公司 | Can automatic recognition cargo position automatic vending machine |
| CN108764778A (en) * | 2018-04-26 | 2018-11-06 | 东华大学 | A kind of warehouse unmanned plane supplement cargo method based on image procossing |
| CN110033452A (en) * | 2019-04-17 | 2019-07-19 | 北京云量互动科技有限公司 | Refrigerator shelf commodity statistics method and system based on image processing techniques |
| CN110245580A (en) * | 2019-05-24 | 2019-09-17 | 北京百度网讯科技有限公司 | A method, apparatus, device and computer storage medium for detecting images |
| CN110472486A (en) * | 2019-07-03 | 2019-11-19 | 北京三快在线科技有限公司 | A kind of shelf obstacle recognition method, device, equipment and readable storage medium storing program for executing |
| CN110610149A (en) * | 2019-09-03 | 2019-12-24 | 卓尔智联(武汉)研究院有限公司 | An information processing method, device and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111161346A (en) | 2020-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111161346B (en) | Method and device for layering commodities in goods shelves and electronic equipment | |
| WO2020048492A1 (en) | Commodity state identification | |
| EP3620981B1 (en) | Object detection method, device, apparatus and computer-readable storage medium | |
| CN110705424B (en) | Method and device for positioning commodity display position and storage medium | |
| US20220262008A1 (en) | Image creation for computer vision model training | |
| US8559726B2 (en) | Image analysis based on pixel brightness grouping | |
| CN104541290A (en) | Methods for providing image feature descriptors | |
| CN111241969A (en) | Target detection method and device and corresponding model training method and device | |
| CN111310706B (en) | Commodity price tag identification method and device, electronic equipment and storage medium | |
| US20100128993A1 (en) | Application of classifiers to sub-sampled integral images for detecting faces in images | |
| CN113627415B (en) | Method and device for determining placement information of target object | |
| CN111160450A (en) | Fruit and vegetable weighing method based on neural network, storage medium and device | |
| US12536685B2 (en) | Image feature matching method, computer device, and storage medium | |
| CN112232334B (en) | Intelligent commodity selling identification and detection method | |
| CN113034574A (en) | Commodity ground heap area calculation method and system based on target detection | |
| JP7130945B2 (en) | Inventory detection program, inventory detection method and inventory detection device | |
| CN110929668A (en) | Commodity detection method and device based on unmanned goods shelf | |
| US20090297055A1 (en) | Supervised edge detection using fractal signatures | |
| US9846949B2 (en) | Determine the shape of a representation of an object | |
| CN110069980A (en) | Product polymerization and device, electronic equipment based on image | |
| CN114299109B (en) | Multi-target object trajectory generation method, system, electronic device and storage medium | |
| CN110533028B (en) | Method and device for detecting commodity display state, electronic device and storage medium | |
| CN115272446B (en) | Method and system for calculating head-stacking occupied area | |
| US11875457B2 (en) | 3D product reconstruction from multiple images collected at checkout lanes | |
| JP6403207B2 (en) | Information terminal equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |