WO2024027854A1 - 快件数量确定方法、装置、电子设备及存储介质 - Google Patents

快件数量确定方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2024027854A1
WO2024027854A1 PCT/CN2023/122075 CN2023122075W WO2024027854A1 WO 2024027854 A1 WO2024027854 A1 WO 2024027854A1 CN 2023122075 W CN2023122075 W CN 2023122075W WO 2024027854 A1 WO2024027854 A1 WO 2024027854A1
Authority
WO
WIPO (PCT)
Prior art keywords
express
image
panoramic
target
shipment
Prior art date
Application number
PCT/CN2023/122075
Other languages
English (en)
French (fr)
Inventor
熊军
Original Assignee
顺丰科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 顺丰科技有限公司 filed Critical 顺丰科技有限公司
Publication of WO2024027854A1 publication Critical patent/WO2024027854A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10141—Special mode during image acquisition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30242—Counting objects in image

Definitions

  • This application relates to the field of express delivery detection, and specifically relates to a method, device, electronic equipment and storage medium for determining the quantity of express delivery.
  • Duplicate inspection is to recover freight for packages with abnormal weight. Before carrying out freight recovery, it is necessary to confirm the result of duplicate weight. During the process of confirming duplicate weight, there is only the current and only package on the scale. From a visual point of view, Determine whether the object in the scale is a single piece. However, the current identification effect of determining whether a package is a single piece or multiple pieces is not good.
  • This application aims to provide a method, device, electronic device and storage medium for determining the quantity of express items, aiming to solve the problem of low accuracy of the method for determining the quantity of express items in the prior art.
  • this application provides a method for determining the quantity of express shipments.
  • the method includes: obtaining a panoramic express shipment image to be identified; determining whether the panoramic express shipment image is an overexposed image; if the panoramic express shipment image is not an overexposed image, Identify and locate the express delivery target in the panoramic express delivery picture; perform quantity identification on the express delivery target to determine the number of express delivery items corresponding to the express delivery target.
  • determining whether the panoramic express image is an overexposed image includes: identifying the panoramic express image using a preset highlight image recognition model to determine whether the panoramic express image is a highlight image; If the panoramic express image is not a highlight image, it is determined that the panoramic express image is not an overexposed image.
  • the method further includes: if the panoramic express image is a highlight image, confirming that the panoramic express image includes multiple overlapping express items.
  • identifying and locating the express target in the panoramic express image includes: using a preset express recognition model to perform analysis on the panoramic express image. Identify and obtain at least one target frame including the express mail; if the panoramic express mail picture includes a conveyor device image, determine the conveyor area corresponding to the conveyor device in the panoramic express mail picture; identify and identify based on the target frame and the conveyor area Locate the shipment item in the panoramic shipment picture mark.
  • identifying and locating the express target in the panoramic express picture according to the target frame and the transfer area includes: determining the distance between the target frame and the transfer area. Whether there is overlap; if there is an overlap between the target frame and the transmission area, determine the location of the target frame as the location of the express delivery target, and obtain the express delivery target in the panoramic express delivery picture.
  • the quantity identification of the express delivery target to determine the express delivery quantity corresponding to the express delivery target includes: based on the position of the target frame in the panoramic express delivery picture, the Crop the panoramic express image to obtain an express image, which is the image to be identified that only includes the express target; obtain the gradient map corresponding to the express image; input the express image and the gradient map into the preset express In the quantity identification model, the shipment quantity corresponding to the shipment target is determined.
  • inputting the shipment picture and the gradient map into a preset shipment quantity identification model to determine the shipment quantity corresponding to the shipment target includes: inputting the shipment picture and the gradient map The image is composed of a four-channel picture, and the four-channel picture is input into the preset shipment quantity identification model to determine the shipment quantity corresponding to the shipment target.
  • the present application provides a device for determining the number of express shipments.
  • the device includes: an acquisition module for obtaining a panoramic express shipment image to be identified; and an overexposure determination module for determining whether the panoramic express shipment image is an overexposed image. ; Identification and positioning module, used to identify and locate the express target in the panoramic express image if the panoramic express image is not an overexposed image; Quantity determination module, used to perform quantitative identification of the express target to determine all The number of shipments corresponding to the express shipment target.
  • the overexposure determination module is configured to: identify the panoramic express image using a preset highlight image recognition model to determine whether the panoramic express image is a highlight image; if the panoramic express image is not If it is a highlight image, it is determined that the panoramic express image is not an overexposed image.
  • the overexposure determination module is also configured to: if the panoramic express image is a highlight image, confirm that the panoramic express image includes multiple overlapping express items.
  • the recognition and positioning module is configured to: use a preset express recognition model to identify the panoramic express image to obtain at least one target frame including the express; if the panoramic express image includes a conveyor device image , determine the transmission area corresponding to the transmission device in the panoramic express picture; identify and locate the express target in the panoramic express picture according to the target frame and the transmission area.
  • the identification and positioning module is configured to: determine whether there is an overlap between the target frame and the transmission area; if there is an overlap between the target frame and the transmission area, determine whether the target frame overlaps with the transmission area.
  • the location of the target frame is the location of the express delivery target, and the express delivery target in the panoramic express delivery picture is obtained.
  • the quantity determination module is configured to: crop the panoramic express image according to the position of the target frame in the panoramic express image to obtain an express image, where the express image is To be identified, only pictures including express delivery targets are obtained; gradient images corresponding to the express delivery images are obtained; and the express delivery images and gradient images are input into a preset express delivery quantity identification model to determine the express delivery quantity corresponding to the express delivery targets.
  • the present application also provides an electronic device, the electronic device includes: one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the in the memory, and configured to be executed by the processor to implement the shipment quantity determining method described in any one of the first aspects.
  • the present application also provides a computer-readable storage medium on which a computer program is stored, and the computer program is loaded by the processor to execute the method for determining the quantity of shipments described in any one of the first aspects. step.
  • This application provides a method, device, electronic equipment and storage medium for determining the quantity of express shipments.
  • the method includes: obtaining a panoramic express image to be identified; determining whether the panoramic express image is an overexposed image; if the panoramic express image is not an overexposed image, identifying and locating the express target in the panoramic express image; and identifying the number of express items corresponding to the express target. .
  • This method identifies the panoramic express image, and first determines whether the panoramic express image is an overexposed image; if it is not an overexposed image, the shipment is then identified and the quantity of the shipment is determined.
  • This application first identifies panoramic express delivery images to avoid missing features of the express delivery close to the edge, resulting in unrecognizability; and removes overexposed images, and only recognizes panoramic express delivery images that are not overexposed, improving the efficiency of express delivery recognition; at the same time, using panoramic express delivery images - overexposed images
  • the triple recognition of exposure recognition-identification positioning-quantity recognition can improve the accuracy of identifying the quantity of express items.
  • FIG. 1 is a schematic diagram of a scenario of an express shipment quantity determination system provided by an embodiment of the present application.
  • Figure 2 is a schematic flowchart of a method for determining the quantity of express shipments provided by an embodiment of the present application.
  • FIG. 3 is a schematic flowchart of determining whether a panoramic express image is an overexposed image according to an embodiment of the present application.
  • Figure 4 is a schematic flowchart of identifying and locating express items provided by an embodiment of the present application.
  • Figure 5 is a schematic flowchart of determining the quantity of express items provided by an embodiment of the present application.
  • Figure 6 is a schematic diagram of a complete process for determining the quantity of express items provided by an embodiment of the present application.
  • Figure 7 is a schematic structural diagram of a device for determining the quantity of express items provided by an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • first and second are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Therefore, the features defined as “first” and “second” may explicitly or implicitly include one or more features. In the description of this application, “plurality” means two or more than two, unless otherwise explicitly and specifically limited.
  • Embodiments of the present application provide a method, device, electronic device, and storage medium for determining the quantity of express shipments, which will be described in detail below.
  • Figure 1 is a schematic diagram of a scenario of an express delivery quantity determination system provided by an embodiment of the present application.
  • the express delivery quantity determination system may include an electronic device 100, and an express delivery quantity determination device is integrated in the electronic device 100, as shown in Figure 1 electronic equipment.
  • the electronic device 100 can be an independent server, or a server network or server cluster composed of servers.
  • the electronic device 100 described in the embodiment of the present application includes, but is not limited to, a computer, a network A host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers.
  • the cloud server consists of a large number of computers or network servers based on cloud computing (Cloud Computing).
  • the application environment shown in Figure 1 is only one application scenario of the solution of the present application and does not constitute a limitation on the application scenarios of the solution of the present application.
  • Other application environments may also include those shown in Figure 1 More or less electronic devices are shown.
  • the express shipment quantity determination system may also include one or more other servers, which are not limited here.
  • the express shipment quantity determination system may also include a storage unit 200 for storing data, such as captured panoramic shipment pictures.
  • scenario diagram of the express delivery quantity determination system shown in Figure 1 is only an example.
  • the express delivery quantity determination system and the scenario described in the embodiment of the present application are for the purpose of more clearly illustrating the technical solution of the embodiment of the present application and are not This constitutes a limitation on the technical solutions provided by the embodiments of the present application.
  • Persons of ordinary skill in the art know that with the evolution of the express shipment quantity determination system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present application can also solve similar technical problems. Be applicable.
  • an embodiment of the present application provides a method for determining the quantity of express items.
  • the execution subject of the method for determining the quantity of express items is an apparatus for determining the quantity of express items.
  • the apparatus for determining the quantity of express items is applied to electronic equipment.
  • the method for determining the quantity of express items includes: obtaining the number to be identified. Panoramic express image; determine whether the panoramic express image is an overexposed image; if the panoramic express image is not an overexposed image, identify and locate the express items in the panoramic express image, and obtain the express target in the panoramic express image; identify the quantity of the express target, To determine the number of shipments corresponding to the shipment target in the panoramic shipment picture.
  • the express shipment quantity determination method includes the following steps.
  • the panoramic express picture corresponding to the conveyor it is necessary to obtain the panoramic express picture corresponding to the conveyor during the transportation of the express on the conveyor, so as to identify whether the express in the panoramic express picture is one express or multiple express.
  • the panoramic express picture is used as the picture to be identified, rather than the picture including only the conveying device as the picture to be identified, because: the panoramic express picture contains the most information, which can avoid the express being at the edge of the conveying device. This leads to the loss of at least part of the characteristics of the express item, and the problem of inaccurate identification due to only part of the express item falling on the conveyor device.
  • panoramic express image is not an overexposed image, identify and locate the express target in the panoramic express image.
  • a camera or other equipment is used to obtain a panoramic express image.
  • the panoramic express image is overexposed, that is, the specific texture of the express cannot be clearly seen. Therefore, it is first necessary to determine whether the panoramic express image is overexposed and whether the texture of the express cannot be determined.
  • the panoramic express image is not an overexposed image, you can proceed to the subsequent steps of determining the number of shipments; if the panoramic express image is an overexposed image, the default panoramic express image includes multiple overlapping shipments, so the panoramic express image needs to be The shipments in the shipment will be re-checked to avoid problems such as abnormal weight of the shipments caused by overlapping shipments.
  • the express delivery recognition model is usually used to identify and locate the express delivery in the panoramic express delivery picture, obtain the express delivery target, and then perform further quantitative judgment on the identified express delivery targets.
  • a shipment target can include one shipment, or multiple overlapping shipments.
  • a panoramic shipment picture includes multiple shipments being transported, but due to the problem of stacking of shipments, some shipments are blocked and cannot be recognized. Therefore, in the embodiment of the present application, it is mainly to determine the actual number of shipments corresponding to the identified shipment targets in the panoramic shipment picture, rather than to determine whether there is one or multiple shipments in the entire panoramic shipment picture.
  • a preset shipment quantity prediction model (or shipment quantity identification model) can be used to determine the quantity of the identified shipments, so that further shipment quantity judgment can be performed based on the identification of the shipments.
  • a preset shipment quantity prediction model can be used to determine the shipment quantity corresponding to the identified shipment target.
  • the method for determining the number of express items includes: obtaining a panoramic express image to be identified; determining whether the panoramic express image is an overexposed image; if the panoramic express image is not an overexposed image, identifying and locating the express item in the panoramic express image Target; perform quantity identification on the express target to determine the number of express items corresponding to the express target.
  • This method identifies the panoramic express image, and first determines whether the panoramic express image is an overexposed image; if it is not an overexposed image, the shipment is then identified and the quantity of the shipment is determined.
  • This application first identifies panoramic express delivery images to avoid missing features of the express delivery close to the edge, resulting in unrecognizability; and removes overexposed images, and only recognizes panoramic express delivery images that are not overexposed to improve the efficiency of express delivery recognition; at the same time, use panoramic express delivery
  • the triple recognition of image - overexposure recognition - recognition positioning - quantity recognition can improve the accuracy of identifying the shipment quantity corresponding to a single shipment target.
  • the determination process may specifically include the following content.
  • panoramic express image is not a highlight image, make sure the panoramic express image is not an overexposed image.
  • a two-class highlight image recognition model can be used to determine whether the panoramic express image is an overexposed image. Specifically, it is determined whether the panoramic express image is a highlight image. If it is a highlight image, the panoramic express image is an overexposed image; if it is not a highlight image, the panoramic express image is not an overexposed image.
  • the highlight image recognition model can be an EfficientNet network, that is, a powerful convolutional neural network, whose network structure is obtained through a composite model expansion method combined with neural structure search technology.
  • the output of the highlight image recognition model can be 0 or 1; among them, 0 means that the panoramic express image is a normal and not overexposed image, and 1 means that the panoramic express image is an overexposed image.
  • the highlight image recognition model can also be other two-classification models, which are not limited in the embodiments of this application.
  • the panoramic express image is a highlight image
  • the panoramic express image is an overexposed image
  • the default overexposed image includes multiple overlapping express items; that is, the default panoramic express image includes multiple overlapping express items.
  • Figure 4 is a schematic flowchart of identifying and locating the express package provided by an embodiment of the present application.
  • the process of identifying and locating the express package may include the following content.
  • each target box can include one express item, and multiple express items correspond to multiple target boxes.
  • the process of identifying express items is actually a process of constantly adjusting the position and size of the target frame.
  • the panoramic express image includes a conveyor device image, determine the conveyor area corresponding to the conveyor device in the panoramic express image.
  • the panoramic express picture obtained in the embodiment of the present application includes other structures in addition to the conveyor, so it is also necessary to Determine the transmission area corresponding to the transmission device in the panoramic shipment picture to determine the shipment currently located in the transmission area.
  • the express target in the panoramic express picture can be identified and positioned.
  • the specific process of identifying and locating the express delivery target please refer to steps 43 to 44 below.
  • the delivery area is individually cut out to determine the express delivery located in the delivery area. Specifically, in the embodiment of the present application, it is determined whether the target frame including the express package overlaps with the transmission area, thereby determining whether the express package is in the transmission area.
  • the location of the target frame is the location of the express package.
  • the target frame is usually the smallest circumscribed rectangle corresponding to the express shipment; and in the embodiment of the present application, determining whether the target frame overlaps with the transmission area mainly depends on the determination of the four vertex corners of the target frame and the center point of the target frame. Whether any of the points falls within the teleportation area, including falling on the edge of the teleportation area.
  • Figure 5 is a schematic flowchart of determining the quantity of express items provided by an embodiment of the present application.
  • the process of determining the quantity of express items (step 24) may include the following content.
  • the express image is a picture to be identified that only includes the express target.
  • the shipment target When determining the number of shipments corresponding to the shipment target, no other image information other than the shipment target is required, so the shipment target can be individually cut out from the panoramic shipment image.
  • the position of the express target in the panoramic express image is actually determined. Therefore, the panoramic express image can be cropped to obtain an express image that only includes the express target.
  • the input of the neural network model is a regular picture, so in this application, the cropping of the panoramic express picture is also based on the size of the target frame.
  • the cropped express picture can be a relatively regular rectangular picture, such as A rectangular image or a square image; rather than being cut to fit the exact outline of the shipment.
  • the gradient map corresponding to the express image can be further determined; the gradient map of the express is mainly to obtain the outline of the express to be identified, so as to determine the express to be identified (express target) based on the outline of the express to be identified ) in the shipment quantity.
  • the specific method of determining the gradient map corresponding to the express image can refer to the existing technology, and is not limited here.
  • the specific process of using the preset shipment quantity identification model to identify the shipment quantity can be found in step 53 below.
  • the main steps are: determine the RGB three-channel image corresponding to the shipment image, and combine the RGB three-channel image and the gradient image to form a four-channel image, so that the four-channel image can be input into the preset shipment quantity identification model to determine the number of shipments in the shipment to be identified. .
  • the gradient map corresponding to the express image is obtained in the embodiment of the present application; the single-channel gradient map and the three-channel RGB map are combined to form a four-channel map.
  • the four-channel graph serves as the input of the express shipment quantity identification model, assisting the model in distinguishing the characteristic information of single express shipments and multiple express shipments, and realizing the identification of the shipment quantity.
  • the shipment quantity identification model can also be a two-classification model, with the output being 0 or 1; 0 means that there is only one shipment to be identified, and there is no overlapping shipment; 1 means that the shipment to be identified is formed by multiple overlapping shipments. .
  • the shipment quantity identification model may be a multi-classification model, and shipment pictures and gradient images may be used as inputs to the shipment quantity identification model, or a four-channel image composed of a single-channel gradient image and a three-channel RGB image may be used.
  • the output of the shipment quantity identification model can be a variety of values, such as 0, 1, 2, 3, etc.
  • the output of the shipment quantity identification model can represent the shipment quantity corresponding to the shipment target.
  • FIG. 6 is a schematic diagram of a complete process for determining the quantity of express items provided by an embodiment of the present application.
  • the process of determining the quantity of express items may include the following contents.
  • step 62 is performed. If the panoramic express image is a highlight image, step 67 is executed to output the recognition result that the express target is multiple express items, that is, the express target includes multiple overlapping express items.
  • the output target is the identification result of a single shipment or multiple shipments.
  • FIG. 7 is a schematic structural diagram of an express delivery quantity determining device provided by an embodiment of the present application.
  • the express delivery quantity determining device may include:
  • the acquisition module 701 is used to acquire the panoramic express picture to be identified.
  • the overexposure determination module 702 is used to determine whether the panoramic express image is an overexposed image.
  • the identification and positioning module 703 is used to identify and locate the express target in the panoramic express image if the panoramic express image is not an overexposed image.
  • the quantity determination module 704 is used to perform quantity identification on the express delivery target to determine the express delivery quantity corresponding to the express delivery target.
  • the express delivery quantity determination device determines whether the panoramic express delivery image is an overexposed image. If the panoramic express delivery image is not an overexposed image, it identifies and locates the express delivery target in the panoramic express delivery image, thereby identifying the corresponding express delivery target. Shipment quantity. This method identifies the panoramic express image, and first determines whether the panoramic express image is an overexposed image; if it is not an overexposed image, the shipment is then identified and the quantity of the shipment is determined.
  • This application first identifies panoramic express delivery images to avoid missing features of the express delivery close to the edge, resulting in unrecognizability; and removes overexposed images, and only recognizes panoramic express delivery images that are not overexposed, improving the efficiency of express delivery recognition; at the same time, using panoramic express delivery images - overexposed images
  • the triple recognition of exposure recognition-identification positioning-quantity recognition can improve the accuracy of identifying the quantity of express items.
  • the overexposure determination module 702 can be specifically configured to: use a preset highlight image recognition model to identify panoramic express images to determine whether the panoramic express image is a highlight image; if the panoramic express image is not a highlight image, determine whether the panoramic express image is a highlight image. Shipment pictures are not overexposed pictures.
  • the overexposure determination module 702 is also used to confirm that the panoramic express image includes multiple overlapping express items if the panoramic express image is a highlight image.
  • the recognition and positioning module 703 can be specifically configured to: use a preset express recognition model to identify the panoramic express image to obtain at least one target frame including the express; if the panoramic express image includes a conveyor device image, determine the panoramic The transmission area corresponding to the transmission device in the express picture; identify and locate the express target in the panoramic express picture according to the target frame and the transmission area.
  • the identification and positioning module 703 can be specifically used to: determine whether there is an overlap between the target frame and the transmission area; if there is an overlap between the target frame and the transmission area, determine that the location of the target frame is the express shipment. The location of the target is obtained to obtain the express target in the panoramic express picture.
  • the quantity determination module 704 can be specifically configured to: crop the panoramic express image according to the position of the target frame in the panoramic express image to obtain an express image, where the express image is a picture to be identified that only includes the express target. ; Obtain the gradient map corresponding to the shipment image; input the shipment image and gradient map into the preset shipment quantity identification model to determine the shipment quantity corresponding to the shipment target.
  • the quantity determination module 704 can be specifically configured to: combine the shipment image and the gradient map into a four-channel image, and input the four-channel image into a preset shipment quantity identification model to determine the shipment quantity corresponding to the shipment target.
  • An embodiment of the present application also provides an electronic device that integrates any of the express delivery quantity determining devices provided by the embodiments of the present application. As shown in Figure 8, it shows a schematic structural diagram of an electronic device provided by an embodiment of the present application. Specifically:
  • the electronic device may include components such as a processor 801 of one or more processing cores, a memory 802 of one or more computer-readable storage media, a power supply 803 and an input unit 804.
  • a processor 801 of one or more processing cores a memory 802 of one or more computer-readable storage media
  • a power supply 803 a power supply 803
  • the electronic device may include more or fewer components than shown in the figures, or combine certain components, or have different components. layout. in:
  • the processor 801 is the control center of the electronic device, using various interfaces and lines to connect various parts of the entire electronic device, by running or executing software programs and/or modules stored in the memory 802, and calling software programs stored in the memory 802. Data, perform various functions of the electronic device and process the data to conduct overall monitoring of the electronic device.
  • the processor 801 may include one or more processing cores; preferably, the processor 801 may integrate an application processor and a modem processor, where the application processor mainly processes operating systems, user interfaces, application programs, etc. , the modem processor mainly handles wireless communications. It can be understood that the above modem processor may not be integrated into the processor 801.
  • the memory 802 can be used to store software programs and modules, and the processor 801 executes various functional applications and data processing by running the software programs and modules stored in the memory 802 .
  • the memory 802 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the storage data area may store data according to Data created by the use of electronic devices, etc.
  • memory 802 may include high-speed random access Access memory may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802 .
  • the electronic device also includes a power supply 803 that supplies power to various components.
  • the power supply 803 can be logically connected to the processor 801 through a power management system, so that functions such as charging, discharging, and power consumption management can be implemented through the power management system.
  • the power supply 803 may also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.
  • the electronic device may also include an input unit 804 that may be used to receive input numeric or character information and generate keyboard, mouse, joystick, optical, or trackball signal input related to user settings and function control.
  • an input unit 804 may be used to receive input numeric or character information and generate keyboard, mouse, joystick, optical, or trackball signal input related to user settings and function control.
  • the electronic device may also include a display unit and the like, which will not be described again here.
  • the processor 801 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 802 according to the following instructions, and the processor 801 will run the executable files stored in The application program in the memory 802 implements various functions, as follows:
  • embodiments of the present application provide a computer-readable storage medium, which may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, etc. .
  • a computer program is stored on the storage medium, and the computer program is loaded by the processor to execute the steps in any of the express shipment quantity determination methods provided by the embodiments of the present application.
  • a computer program loaded by a processor may perform the following steps:
  • each of the above units or structures can be implemented as an independent entity, or can be combined in any way and implemented as the same or several entities.
  • each of the above units or structures please refer to the previous method embodiments. Here No longer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Information Transfer Between Computers (AREA)
  • Image Analysis (AREA)

Abstract

一种快件数量确定方法、装置、电子设备及存储介质。该方法包括：获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标；对快件目标进行数量识别，以确定快件目标对应的快件数量。本申请的技术方案通过对未过曝的全景快件图片进行识别，可以提高快件识别的效率，以及提高识别快件数量的精度。

Description

快件数量确定方法、装置、电子设备及存储介质 技术领域
本申请涉及快件检测领域，具体涉及一种快件数量确定方法、装置、电子设备及存储介质。
发明背景
物流快递按重量或体积计算运费，收件端通过人工测量重量或体积容易发生少计等情况，从而带来公司收入减少等诸多不良影响，所以需要复重并根据复重结果追缴运费。
复重稽查是为了对重量异常的包裹进行运费追缴，在进行运费追缴前需要对复重的结果进行确定，确认复重的过程中秤面上只有当前唯一的包裹，对于视觉的角度来说就是判断秤内的物体是否是单件。但目前确定包裹是单件还是多件的识别效果并不好。
发明内容
本申请旨在提供一种快件数量确定方法、装置、电子设备及存储介质，旨在解决现有技术中快件数量确定方法的准确度较低的问题。
一方面，本申请提供一种快件数量确定方法，所述方法包括：获取待识别的全景快件图片；判断所述全景快件图片是否为过曝图片；若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标；对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量。
在一些可能的实施例中，所述判断所述全景快件图片是否为过曝图片，包括：利用预设高光图片识别模型识别所述全景快件图片，以确定所述全景快件图片是否为高光图片；若所述全景快件图片不为高光图片，则确定所述全景快件图片不为过曝图片。
在一些可能的实施例中，所述方法还包括：若所述全景快件图片为高光图片，则确认所述全景快件图片中包括多个重叠快件。
在一些可能的实施例中，所述若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标，包括：利用预设快件识别模型对所述全景快件图片进行识别，得到至少一个包括快件的目标框；若所述全景快件图片包括传送装置图像，确定所述全景快件图片中的传送装置对应的传送区域；根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目 标。
在一些可能的实施例中，所述根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目标，包括：判断所述目标框与所述传送区域之间是否重叠；若所述目标框和所述传送区域之间发生重叠，则确定所述目标框所在的位置为所述快件目标所在的位置，得到所述全景快件图片中的所述快件目标。
在一些可能的实施例中，所述对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量，包括：根据所述目标框在所述全景快件图片中的位置，对所述全景快件图片进行裁切，得到快件图片，所述快件图片为待识别的仅包括快件目标的图片；获取所述快件图片对应的梯度图；将所述快件图片和所述梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量。
在一些可能的实施例中，所述将所述快件图片和所述梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量，包括：将所述快件图片和所述梯度图组成四通道图片，并将所述四通道图片输入所述预设快件数量识别模型中，以确定所述快件目标对应的快件数量。
另一方面，本申请提供一种快件数量确定装置，所述装置包括：获取模块，用于获取待识别的全景快件图片；过曝判断模块，用于判断所述全景快件图片是否为过曝图片；识别定位模块，用于若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标；数量确定模块，用于对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量。
在一些可能的实施例中，所述过曝判断模块用于：利用预设高光图片识别模型识别所述全景快件图片，以确定所述全景快件图片是否为高光图片；若所述全景快件图片不为高光图片，则确定所述全景快件图片不为过曝图片。
在一些可能的实施例中，所述过曝判断模块还用于：若所述全景快件图片为高光图片，则确认所述全景快件图片中包括多个重叠快件。
在一些可能的实施例中，所述识别定位模块用于：利用预设快件识别模型对所述全景快件图片进行识别，得到至少一个包括快件的目标框；若所述全景快件图片包括传送装置图像，确定所述全景快件图片中的传送装置对应的传送区域；根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目标。
在一些可能的实施例中，所述识别定位模块用于：判断所述目标框与所述传送区域之间是否重叠；若所述目标框和所述传送区域之间发生重叠，则确定所述目标框所在的位置为所述快件目标所在的位置，得到所述全景快件图片中的所述快件目标。
在一些可能的实施例中，所述数量确定模块用于：根据所述目标框在所述全景快件图片中的位置，对所述全景快件图片进行裁切，得到快件图片，所述快件图片为待识别的仅包括快件目标的图片；获取所述快件图片对应的梯度图；将所述快件图片和所述梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量。
另一方面，本申请还提供一种电子设备，所述电子设备包括：一个或多个处理器；存储器；以及一个或多个应用程序，其中所述一个或多个应用程序被存储于所述存储器中，并配置为由所述处理器执行以实现第一方面中任一项所述的快件数量确定方法。
另一方面，本申请还提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器进行加载，以执行第一方面任一项所述的快件数量确定方法中的步骤。
本申请提供一种快件数量确定方法、装置、电子设备及存储介质。该方法包括：获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标；识别快件目标对应的快件数量。该方法对全景快件图片进行识别，且首先判断全景快件图片是否为过曝图片；若不是过曝图片，再进行快件的识别和快件数量的确定。本申请首先识别全景快件图片，避免靠近边缘的快件特征缺失导致无法识别；且去除过曝图片，仅对未过曝的全景快件图片进行识别，提高快件识别的效率；同时利用全景快件图片-过曝识别-识别定位-数量识别的三次识别可提高识别快件数量的精度。
附图简要说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请一实施例所提供的快件数量确定系统的场景示意图。
图2是本申请一实施例提供的快件数量确定方法的流程示意图。
图3为本申请一实施例提供的判断全景快件图片是否为过曝图片的流程示意图。
图4为本申请一实施例提供的识别定位快件的流程示意图。
图5为本申请一实施例提供的确定快件数量的流程示意图。
图6为本申请一实施例提供的确定快件数量的完整流程示意图。
图7为本申请一实施例提供的快件数量确定装置的结构示意图。
图8是本申请一实施例提供的电子设备的结构示意图。
实施本发明的方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
在本申请的描述中，需要理解的是，术语“中心”、“纵向”、“横向”、“长度”、“宽度”、“厚度”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”、“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本申请的限制。此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个特征。在本申请的描述中，“多个”的含义是两个或两个以上，除非另有明确具体的限定。
在本申请中，“示例性”一词用来表示“用作例子、例证或说明”。本申请中被描述为“示例性”的任何实施例不一定被解释为比其它实施例更优选或更具优势。为了使本领域任何技术人员能够实现和使用本申请，给出了以下描述。在以下描述中，为了解释的目的而列出了细节。应当明白的是，本领域普通技术人员可以认识到，在不使用这些特定细节的情况下也可以实现本申请。在其它实例中，不会对公知的结构和过程进行详细阐述，以避免不必要的细节使本申请的描述变得晦涩。因此，本申请并非旨在限于所示的实施例，而是与符合本申请所公开的原理和特征的最广范围相一致。
需要说明的是，本申请实施例提供的方法由于是在电子设备中执行，各电子设备的处理对象均以数据或信息的形式存在，例如时间，实质为时间信息，可以理解的是，后续实施例中若提及尺寸、数量、位置等，均为对应的数据存在，以便电子设备进行处理，具体此处不作赘述。
本申请实施例提供一种快件数量确定方法、装置、电子设备及存储介质，以下分别进行详细说明。
请参阅图1，图1为本申请一实施例所提供的快件数量确定系统的场景示意图，该快件数量确定系统可以包括电子设备100，电子设备100中集成有快件数量确定装置，如图1中的电子设备。
本申请实施例中，该电子设备100可以是独立的服务器，也可以是服务器组成的服务器网络或服务器集群，例如，本申请实施例中所描述的电子设备100，其包括但不限于计算机、网络主机、单个网络服务器、多个网络服务器集或多个服务器构成的云服务器。其中，云服务器由基于云计算(Cloud Computing)的大量计算机或网络服务器构成。
本领域技术人员可以理解，图1中示出的应用环境，仅仅是本申请方案的一种应用场景，并不构成对本申请方案应用场景的限定，其他的应用环境还可以包括比图1中所示更多或更少的电子设备，例如图1中仅示出1个电子设备，可以理解的，该快件数量确定系统还可以包括一个或多个其他服务器，具体此处不作限定。
另外，如图1所示，该快件数量确定系统还可以包括存储单元200，用于存储数据，如拍摄的全景快件图片等。
需要说明的是，图1所示的快件数量确定系统的场景示意图仅仅是一个示例，本申请实施例描述的快件数量确定系统以及场景是为了更加清楚的说明本申请实施例的技术方案，并不构成对于本申请实施例提供的技术方案的限定，本领域普通技术人员可知，随着快件数量确定系统的演变和新业务场景的出现，本申请实施例提供的技术方案对于类似的技术问题，同样适用。
首先，本申请实施例中提供一种快件数量确定方法，该快件数量确定方法的执行主体为快件数量确定装置，该快件数量确定装置应用于电子设备，该快件数量确定方法包括：获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件，得到全景快件图片中的快件目标；对快件目标进行数量识别，以确定全景快件图片中的快件目标对应的快件数量。
参阅图2，图2是本申请一实施例提供的快件数量确定方法的流程示意图。如图2所示，该快件数量确定方法包括如下步骤。
21、获取待识别的全景快件图片。
在具体的快件寄递过程中，通常需要对传送装置上的快件进行识别；且由于存在快件堆叠的问题，因此需要判断识别出的快件是单独的一个快件，还是发生重叠的多个快件。
而本申请实施例中需要在快件在传送装置上运输的过程中，获取传送装置对应的全景快件图片，以识别全景快件图片中的快件是一个快件还是多个快件。其中，全景快件图片中的物体通常多于快件本身，即在利用摄像头等设备获取全景图像(全景快件图片)时，需要将摄像头与传送装置之间的距离设定的较大，使得摄像头不仅可以拍摄传送装置，也可以获取传送装置之外的其他物体。
本申请实施例中将全景快件图片作为待识别的图片，而非将仅包括传送装置的图片作为待识别的图片是因为：全景快件图片中含有最多的信息，可以避免由于快件处于传送装置的边缘而导致快件的至少部分特征丢失、以及因快件仅有部分落在传送装置上方而导致的识别不准确的问题。
22、判断全景快件图片是否为过曝图片。
23、若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标。
在本申请的实施例中，是利用摄像头等设备获取全景快件图片，而由于光线和拍摄角度的问题，导致全景快件图片存在过曝情况，即无法看清快件的具体纹理。因此首先需要判断全景快件图片是否为过曝图片，是否存在无法确定快件纹理的情况。
若全景快件图片不为过曝图片，则可以进行后续确定快件数量的步骤；而若全景快件图片为过曝图片，则默认全景快件图片中包括多个重叠的快件，那么需要对该全景快件图片中的快件重新进行检查，避免由于快件重叠导致的快件重量异常等问题。
本申请实施例中通常利用快件识别模型识别并定位出全景快件图片中存在的快件，得到快件目标，再对识别出的快件目标进行进一步的数量判断。例如，快件目标可以包括一个快件，或包括多个重叠的快件。
24、对快件目标进行数量识别，以确定快件目标对应的快件数量。
通常情况下，一张全景快件图片中会包括多个正在传输的快件，但由于存在快件堆叠的问题，导致某些快件被遮挡无法被识别。因此在本申请的实施例中，主要是判断全景快件图片中已经识别出的快件目标对应的实际快件数量，而不是判断整个全景快件图片中存在一个还是多个快件。
在本申请的实施例中，可以利用预设的快件数量预测模型(或快件数量识别模型)对识别出的快件进行数量判断，以在识别出快件的基础上进行进一步的快件数量判断。例如，可以利用预设的快件数量预测模型对识别出的快件目标所对应的快件数量进行判断。
本申请实施例提供的快件数量确定方法，包括：获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标；对快件目标进行数量识别，以确定快件目标对应的快件数量。该方法对全景快件图片进行识别，且首先判断全景快件图片是否为过曝图片；若不是过曝图片，再进行快件的识别和快件数量的确定。本申请首先识别全景快件图片，避免靠近边缘的快件特征缺失导致无法识别；且去除过曝图片，仅对未过曝的全景快件图片进行识别，提高快件识别的效率；同时利用全景快件 图片-过曝识别-识别定位-数量识别的三次识别可提高识别单个快件目标对应的快件数量的精度。
如图3所示，为本申请一实施例提供的判断全景快件图片是否为过曝图片的流程示意图，判断过程(步骤22)具体可以包括如下内容。
31、利用预设高光图片识别模型识别全景快件图片，以确定全景快件图片是否为高光图片。
32、若全景快件图片不为高光图片，则确定全景快件图片不为过曝图片。
在本申请的实施例中，可以利用两分类的高光图片识别模型来判断全景快件图片是否为过曝图片。具体为判断全景快件图片是否高光图片，若为高光图片，则全景快件图片为过曝图片；若不为高光图片，则全景快件图片不为过曝图片。
在一个具体实施例中，高光图片识别模型可以为EfficientNet网络，即，一个性能强大的卷积神经网络，其网络结构是通过复合模型扩张方法结合神经结构搜索技术获得的。此时高光图片识别模型的输出可以为0或1；其中，0代表全景快件图片为正常不过曝的图片，1则代表全景快件图片为过曝图片。在其他实施例中，高光图片识别模型也可以为其他二分类模型，本申请实施例中不做限定。
在上述实施例中，若是识别出全景快件图片为高光图片，则全景快件图片为过曝图片，默认过曝图片中包括多个重叠放置的快件；即默认全景快件图片中包括多个重叠的快件，无需再进行后续的快件识别定位和快件数量确定的步骤。
在确定了非过曝的全景快件图片后，还需要进一步确定快件在全景快件图片中的位置。因此需要利用预设快件识别模型识别并定位全景快件图片中的快件，得到待识别的快件。如图4所示，图4为本申请一实施例提供的识别定位快件的流程示意图，识别定位快件的过程(步骤23)可以包括如下内容。
41、利用预设快件识别模型对全景快件图片进行识别，得到至少一个包括快件的目标框。
在具体的快件识别过程中，会生成多个目标框，而每一个目标框中可包括一个快件，多个快件对应多个目标框。识别快件的过程实际上也是不断调整目标框的位置和尺寸的过程。
需要说明的是，在识别快件时，多个重叠的快件会识别为一个快件，对应一个目标框。
42、若全景快件图片包括传送装置图像，确定全景快件图片中的传送装置对应的传送区域。
在本申请的实施例中，由于快件是在传送装置上进行传送的，而本申请实施例中获取的全景快件图片中除传送装置之外，还包括有其他结构，因此还需要在 全景快件图片中确定传送装置对应的传送区域，以便确定当前位于传送区域上的快件。
根据目标框和传送区域，可识别并定位全景快件图片中的快件目标。识别并定位快件目标的具体过程可参见如下步骤43至44。
43、判断目标框和传送区域之间是否重叠。
44、若目标框和传送区域之间发生重叠，则确定目标框所在的位置为快件目标所在的位置，得到全景快件图片中的快件目标。
由于本申请实施例中首先获取了全景快件图片，因此可以避免遗漏快件；而为了进一步避免遗漏快件，此处又将传送区域单独裁切出来判断位于传送区域上的快件。具体地，本申请实施例中判断包括快件的目标框和传送区域之间是否重叠，以此判断快件是否在传送区域中。
在本申请的实施例中，只要目标框和传送区域之间发生重叠，就认为目标框中存在快件，且目标框所在的位置即为快件的位置。在本申请的实施例中，不需要判断目标框与传送区域之间重叠的区域大小，只要两者之间发生重叠即可。这是因为某些快件可能在传送区域的边缘，而确定传送区域时，将这部分快件超出传送区域部分去除掉了。若是判断目标框与传送区域之间重叠的区域大小，再判断目标框中是否为快件，则可能遗漏这部分快件。
在上述实施例中，目标框通常为快件对应的最小外接矩形；且本申请实施例中判断目标框和传送区域之间是否重叠主要是判断，目标框的四个顶角和目标框的中心点中任一点是否落在传送区域内部，包括落在传送区域边缘上。
在确定了待识别的快件(如快件目标)后，需要进一步确定待识别快件对应的数量。如图5所示，图5为本申请一实施例提供的确定快件数量的流程示意图，确定快件数量的过程(步骤24)可以包括如下内容。
51、根据目标框在全景快件图片中的位置，对全景快件图片进行裁切，得到快件图片，快件图片为待识别的仅包括快件目标的图片。
在确定快件目标对应的快件数量时，不需要除快件目标之外的其他图片信息，因此可以将快件目标从全景快件图片中单独裁切出来。而识别出快件目标的同时，实际上也确定快件目标在全景快件图片中的位置，因此可以实现对全景快件图片的裁切，得到仅包括快件目标的快件图片。
通常来说，神经网络模型的输入为规则的图片，因此本申请中对全景快件图片进行裁切也是根据目标框的尺寸进行裁切，裁切出的快件图片可以是比较规则的矩形图片，如长方形图片或正方形图片；而非是按照快件的具体轮廓进行裁切。
52、获取快件图片对应的梯度图。
在确定了仅包括快件目标的快件图片后，可以进一步确定快件图片对应的梯度图；快件的梯度图主要是为了获取待识别快件的轮廓，以根据待识别快件的轮廓确定待识别快件(快件目标)中的快件数量。确定快件图片对应的梯度图的具体方法可以参考现有技术，此处不做限定。
将快件图片和梯度图输入预设快件数量识别模型中，可确定全景快件图片中的快件目标对应的快件数量。利用预设快件数量识别模型识别快件数量的具体过程可参见如下步骤53。
53、将快件图片和梯度图组成四通道图片，并将四通道图片输入预设快件数量识别模型中，以确定快件目标对应的快件数量。
在本申请的实施例中，不仅需要利用快件图片对应的梯度图，还需要同时利用快件图片本身以识别快件数量。
具体主要为：确定快件图片对应的RGB三通道图，并将RGB三通道图和梯度图组成四通道图片，以将四通道图片输入预设快件数量识别模型中，确定待识别快件中的快件数量。
在上述实施例中，由于梯度图对应的特征比较明显，因此本申请实施例中获取了快件图片对应的梯度图；并将单通道的梯度图和三通道的RGB图结合组成四通道图，将四通道图作为快件数量识别模型的输入，辅助模型区分单件快件和多件快件的特征信息，实现快件数量的识别。
具体地，快件数量识别模型同样可以为一个二分类模型，输出为0或1；0代表待识别快件仅有一个快件，不存在重叠放置的快件；1代表待识别快件为多件快件重叠放置形成。
需要说明的是，在本申请的一些实施例中，仅能判断待识别快件是一件还是多件；当待识别快件为多件时，无法识别具体的快件数量。
在另一些实施例中，快件数量识别模型可以为多分类模型，可以将快件图片和梯度图作为快件数量识别模型的输入，或将单通道的梯度图和三通道的RGB图组成的四通道图作为快件数量识别模型的输入。快件数量识别模型的输出可以是多种数值，如0，1，2，3等，快件数量识别模型的输出可表示快件目标对应的快件数量。
如图6所示，图6为本申请一实施例提供的确定快件数量的完整流程示意图，在该实施例中，确定快件数量的过程可包括如下内容。
61、利用高光图片识别模型对全景快件图片进行高光识别，判断全景快件图片是否为高光图片。
若全景快件图片不为过曝的高光图片，则执行步骤62。若全景快件图片为高光图片，则执行步骤67，输出快件目标是多件快件的识别结果，即快件目标包括多个重叠放置的快件。
62、利用快件识别定位模型识别出全景快件图片中的快件目标，以及定位快件目标所在的位置。
63、对全景快件图片进行裁切，得到快件目标对应的快件图片。
在确定了快件目标的位置后，可以将快件目标从全景快件图片中裁切出来，得到仅包括快件目标的快件图片。
64、获取快件图片对应的RGB三通道图。
65、获取快件图片对应的单通道的梯度图。
66、将RGB三通道图和单通道的梯度图合并为四通道图后输入快件数量识别模型中。
67、通过快件数量识别模型输出快件目标是单件快件或多件快件的识别结果。
本申请实施例还提供一种快件数量确定装置，如图7所示，图7为本申请一实施例提供的快件数量确定装置的结构示意图，该快件数量确定装置可以包括：
获取模块701，用于获取待识别的全景快件图片。
过曝判断模块702，用于判断全景快件图片是否为过曝图片。
识别定位模块703，用于若全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标。
数量确定模块704，用于对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量。
本申请实施例提供的快件数量确定装置，通过判断全景快件图片是否为过曝图片，若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标，从而可识别快件目标对应的快件数量。该方法对全景快件图片进行识别，且首先判断全景快件图片是否为过曝图片；若不是过曝图片，再进行快件的识别和快件数量的确定。本申请首先识别全景快件图片，避免靠近边缘的快件特征缺失导致无法识别；且去除过曝图片，仅对未过曝的全景快件图片进行识别，提高快件识别的效率；同时利用全景快件图片-过曝识别-识别定位-数量识别的三次识别可提高识别快件数量的精度。
在一些实施例中，过曝判断模块702具体可以用于：利用预设高光图片识别模型识别全景快件图片，以确定全景快件图片是否为高光图片；若全景快件图片不为高光图片，则确定全景快件图片不为过曝图片。
在一些实施例中，过曝判断模块702还用于若全景快件图片为高光图片，则确认全景快件图片中包括多个重叠快件。
在一些实施例中，识别定位模块703具体可以用于：利用预设快件识别模型对所述全景快件图片进行识别，得到至少一个包括快件的目标框；若全景快件图片包括传送装置图像，确定全景快件图片中的传送装置对应的传送区域；根据目标框和所述传送区域，识别并定位全景快件图片中的快件目标。
在一些实施例中，识别定位模块703具体可以用于：判断目标框与所述传送区域之间是否重叠；若目标框和所述传送区域之间发生重叠，则确定目标框所在的位置为快件目标所在的位置，得到全景快件图片中的快件目标。
在一些实施例中，数量确定模块704具体可以用于：根据目标框在全景快件图片中的位置，对全景快件图片进行裁切，得到快件图片，快件图片为待识别的仅包括快件目标的图片；获取快件图片对应的梯度图；将快件图片和梯度图输入预设快件数量识别模型中，确定快件目标对应的快件数量。
在一些实施例中，数量确定模块704具体可以用于：将快件图片和梯度图组成四通道图片，并将四通道图片输入预设快件数量识别模型中，以确定快件目标对应的快件数量。
本申请实施例还提供一种电子设备，其集成了本申请实施例所提供的任一种快件数量确定装置。如图8所示，其示出了本申请一实施例提供的电子设备的结构示意图，具体来讲：
该电子设备可以包括一个或者一个以上处理核心的处理器801、一个或一个以上计算机可读存储介质的存储器802、电源803和输入单元804等部件。本领域技术人员可以理解，图中示出的电子设备结构并不构成对电子设备的限定，电子设备可以包括比图示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。其中：
处理器801是该电子设备的控制中心，利用各种接口和线路连接整个电子设备的各个部分，通过运行或执行存储在存储器802内的软件程序和/或模块，以及调用存储在存储器802内的数据，执行电子设备的各种功能和处理数据，从而对电子设备进行整体监控。可选的，处理器801可包括一个或多个处理核心；优选的，处理器801可集成应用处理器和调制解调处理器，其中，应用处理器主要处理操作系统、用户界面和应用程序等，调制解调处理器主要处理无线通信。可以理解的是，上述调制解调处理器也可以不集成到处理器801中。
存储器802可用于存储软件程序以及模块，处理器801通过运行存储在存储器802的软件程序以及模块，从而执行各种功能应用以及数据处理。存储器802可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据电子设备的使用所创建的数据等。此外，存储器802可以包括高速随机 存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地，存储器802还可以包括存储器控制器，以提供处理器801对存储器802的访问。
电子设备还包括给各个部件供电的电源803，优选的，电源803可以通过电源管理系统与处理器801逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。电源803还可以包括一个或一个以上的直流或交流电源、再充电系统、电源故障检测电路、电源转换器或者逆变器、电源状态指示器等任意组件。
该电子设备还可包括输入单元804，该输入单元804可用于接收输入的数字或字符信息，以及产生与用户设置以及功能控制有关的键盘、鼠标、操作杆、光学或者轨迹球信号输入。
尽管未示出，电子设备还可以包括显示单元等，在此不再赘述。具体在本实施例中，电子设备中的处理器801会按照如下的指令，将一个或一个以上的应用程序的进程对应的可执行文件加载到存储器802中，并由处理器801来运行存储在存储器802中的应用程序，从而实现各种功能，如下：
获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标；对快件目标进行数量识别，以确定快件目标对应的快件数量。
本领域普通技术人员可以理解，上述实施例的各种方法中的全部或部分步骤可以通过指令来完成，或通过指令控制相关的硬件来完成，该指令可以存储于一计算机可读存储介质中，并由处理器进行加载和执行。
为此，本申请实施例提供一种计算机可读存储介质，该存储介质可以包括：只读存储器(ROM，Read Only Memory)、随机存取记忆体(RAM，Random Access Memory)、磁盘或光盘等。存储介质上存储有计算机程序，计算机程序被处理器进行加载，以执行本申请实施例所提供的任一种快件数量确定方法中的步骤。例如，计算机程序被处理器进行加载可以执行如下步骤：
获取待识别的全景快件图片；判断全景快件图片是否为过曝图片；若全景快件图片不为过曝图片，识别并定位全景快件图片中的快件目标；对快件目标进行数量识别，以确定快件目标对应的快件数量。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见上文针对其他实施例的详细描述，此处不再赘述。
具体实施时，以上各个单元或结构可以作为独立的实体来实现，也可以进行任意组合，作为同一或若干个实体来实现，以上各个单元或结构的具体实施可参见前面的方法实施例，在此不再赘述。
以上各个操作的具体实施可参见前面的实施例，在此不再赘述。
以上对本申请实施例所提供的一种快件数量确定方法、装置、电子设备及存储介质进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想；同时，对于本领域的技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上，本说明书内容不应理解为对本申请的限制。

Claims (15)

  1. 一种快件数量确定方法，其特征在于，所述方法包括：
    获取待识别的全景快件图片；
    判断所述全景快件图片是否为过曝图片；
    若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标；
    对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量。
  2. 根据权利要求1所述的快件数量确定方法，其特征在于，所述判断所述全景快件图片是否为过曝图片，包括：
    利用预设高光图片识别模型识别所述全景快件图片，以确定所述全景快件图片是否为高光图片；
    若所述全景快件图片不为高光图片，则确定所述全景快件图片不为过曝图片。
  3. 根据权利要求2所述的快件数量确定方法，其特征在于，所述方法还包括：
    若所述全景快件图片为高光图片，则确认所述全景快件图片中包括多个重叠快件。
  4. 根据权利要求1至3中任一项所述的快件数量确定方法，其特征在于，所述若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标，包括：
    利用预设快件识别模型对所述全景快件图片进行识别，得到至少一个包括快件的目标框；
    若所述全景快件图片包括传送装置图像，确定所述全景快件图片中的传送装置对应的传送区域；
    根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目标。
  5. 根据权利要求4所述的快件数量确定方法，其特征在于，所述根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目标，包括：
    判断所述目标框与所述传送区域之间是否重叠；
    若所述目标框和所述传送区域之间发生重叠，则确定所述目标框所在的位置为所述快件目标所在的位置，得到所述全景快件图片中的所述快件目标。
  6. 根据权利要求4或5所述的快件数量确定方法，其特征在于，所述对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量，包括：
    根据所述目标框在所述全景快件图片中的位置，对所述全景快件图片进行裁切，得到快件图片，所述快件图片为待识别的仅包括快件目标的图片；
    获取所述快件图片对应的梯度图；
    将所述快件图片和所述梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量。
  7. 根据权利要求6所述的快件数量确定方法，其特征在于，所述将所述快件图片和所述梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量，包括：
    将所述快件图片和所述梯度图组成四通道图片，并将所述四通道图片输入所述预设快件数量识别模型中，以确定所述快件目标对应的快件数量。
  8. 一种快件数量确定装置，其特征在于，所述装置包括：
    获取模块，用于获取待识别的全景快件图片；
    过曝判断模块，用于判断所述全景快件图片是否为过曝图片；
    识别定位模块，用于若所述全景快件图片不为过曝图片，识别并定位所述全景快件图片中的快件目标；
    数量确定模块，用于对所述快件目标进行数量识别，以确定所述快件目标对应的快件数量。
  9. 根据权利要求8所述的快件数量确定装置，其特征在于，所述过曝判断模块用于：
    利用预设高光图片识别模型识别所述全景快件图片，以确定所述全景快件图片是否为高光图片；
    若所述全景快件图片不为高光图片，则确定所述全景快件图片不为过曝图片。
  10. 根据权利要求9所述的快件数量确定装置，其特征在于，所述过曝判断模块还用于：
    若所述全景快件图片为高光图片，则确认所述全景快件图片中包括多个重叠快件。
  11. 根据权利要求8至10中任一项所述的快件数量确定装置，其特征在于，所述识别定位模块用于：
    利用预设快件识别模型对所述全景快件图片进行识别，得到至少一个包括快件的目标框；
    若所述全景快件图片包括传送装置图像，确定所述全景快件图片中的传送装置对应的传送区域；
    根据所述目标框和所述传送区域，识别并定位所述全景快件图片中的所述快件目标。
  12. 根据权利要求11所述的快件数量确定装置，其特征在于，所述识别定位模块用于：
    判断所述目标框与所述传送区域之间是否重叠；
    若所述目标框和所述传送区域之间发生重叠，则确定所述目标框所在的位置为所述快件目标所在的位置，得到所述全景快件图片中的所述快件目标。
  13. 根据权利要求11或12所述的快件数量确定装置，其特征在于，所述数量确定模块用于：
    根据所述目标框在所述全景快件图片中的位置，对所述全景快件图片进行裁切，得到快件图片，所述快件图片为待识别的仅包括快件目标的图片；
    获取所述快件图片对应的梯度图；
    将所述快件图片和梯度图输入预设快件数量识别模型中，确定所述快件目标对应的快件数量。
  14. 一种电子设备，其特征在于，所述电子设备包括：
    一个或多个处理器；
    存储器；以及
    一个或多个应用程序，其中所述一个或多个应用程序被存储于所述存储器中，并配置为由所述处理器执行以实现权利要求1至7任一项所述的快件数量确定方法。
  15. 一种计算机可读存储介质，其特征在于，其上存储有计算机程序，所述计算机程序被处理器进行加载，以执行权利要求1至7任一项所述的快件数量确定方法中的步骤。
PCT/CN2023/122075 2022-08-05 2023-09-27 快件数量确定方法、装置、电子设备及存储介质 WO2024027854A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210940120.4 2022-08-05
CN202210940120.4A CN117557896A (zh) 2022-08-05 2022-08-05 快件数量确定方法、装置、电子设备及存储介质

Publications (1)

Publication Number Publication Date
WO2024027854A1 true WO2024027854A1 (zh) 2024-02-08

Family

ID=89821085

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/122075 WO2024027854A1 (zh) 2022-08-05 2023-09-27 快件数量确定方法、装置、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN117557896A (zh)
WO (1) WO2024027854A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109376656A (zh) * 2018-10-24 2019-02-22 弭吉荣 一种通过拍照方式分辨物品种类并计算数量的方法和系统
CN109740606A (zh) * 2018-12-20 2019-05-10 上海众源网络有限公司 一种图像识别方法及装置
CN114612739A (zh) * 2022-02-24 2022-06-10 江西裕丰智能农业科技有限公司 双目全景图像目标检测方法、装置、计算机设备
CN114693039A (zh) * 2020-12-30 2022-07-01 顺丰科技有限公司 异常快件识别方法、装置、计算机设备和存储介质
CN114781527A (zh) * 2022-04-27 2022-07-22 武汉大学中南医院 一种基于荧光照片的核型识别方法、装置及电子设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109376656A (zh) * 2018-10-24 2019-02-22 弭吉荣 一种通过拍照方式分辨物品种类并计算数量的方法和系统
CN109740606A (zh) * 2018-12-20 2019-05-10 上海众源网络有限公司 一种图像识别方法及装置
CN114693039A (zh) * 2020-12-30 2022-07-01 顺丰科技有限公司 异常快件识别方法、装置、计算机设备和存储介质
CN114612739A (zh) * 2022-02-24 2022-06-10 江西裕丰智能农业科技有限公司 双目全景图像目标检测方法、装置、计算机设备
CN114781527A (zh) * 2022-04-27 2022-07-22 武汉大学中南医院 一种基于荧光照片的核型识别方法、装置及电子设备

Also Published As

Publication number Publication date
CN117557896A (zh) 2024-02-13

Similar Documents

Publication Publication Date Title
US20210374472A1 (en) Systems and methods for creating training data
WO2021136509A1 (zh) 检测包裹的方法、装置、计算设备、物流系统及存储介质
JP2017171448A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
JP6558020B2 (ja) 検品処理装置、検品処理方法及びプログラム
WO2019153508A1 (zh) 一种行李的管理方法及其设备
WO2023236825A1 (zh) 容积使用率的监控方法、装置和计算机可读存储介质
JP2017171444A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
JP2017171443A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
WO2024061101A1 (zh) 非常规件识别方法、装置、计算机设备及存储介质
CN114170442A (zh) 机器人空间抓取点的确定方法及装置
JP2018036770A (ja) 位置姿勢推定装置、位置姿勢推定方法、及び位置姿勢推定プログラム
CN112756265A (zh) 物品分拣异常检测方法、装置工控设备及存储介质
JP2017171445A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
WO2024027854A1 (zh) 快件数量确定方法、装置、电子设备及存储介质
WO2024140083A1 (zh) 装载率测量方法、装置、设备及介质
CN114140730A (zh) 目标匹配方法、装置、设备及存储介质
CN115471439A (zh) 显示面板缺陷的识别方法、装置、电子设备及存储介质
CN114529843A (zh) 货物拥堵识别方法、装置、电子设备和存储介质
US20220128347A1 (en) System and method to measure object dimension using stereo vision
CN112990189B (zh) 货物拥堵识别方法、装置、电子设备及存储介质
CN115619698A (zh) 电路板缺陷的检测方法、装置及模型训练方法
WO2024088251A1 (zh) 包裹尺寸测量方法、装置、计算机设备及存储介质
JP2020114778A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
JP2017171447A (ja) 投影指示装置、荷物仕分けシステムおよび投影指示方法
CN113496142A (zh) 物流件的体积测量方法以及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23849558

Country of ref document: EP

Kind code of ref document: A1