WO2019218152A1 - 一种智能售货柜 - Google Patents

一种智能售货柜 Download PDF

Info

Publication number
WO2019218152A1
WO2019218152A1 PCT/CN2018/086876 CN2018086876W WO2019218152A1 WO 2019218152 A1 WO2019218152 A1 WO 2019218152A1 CN 2018086876 W CN2018086876 W CN 2018086876W WO 2019218152 A1 WO2019218152 A1 WO 2019218152A1
Authority
WO
WIPO (PCT)
Prior art keywords
layer
camera
height
partition
smart vending
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.)
Ceased
Application number
PCT/CN2018/086876
Other languages
English (en)
French (fr)
Inventor
张站朝
李强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Cloudminds Shenzhen Robotics Systems Co Ltd
Original Assignee
Cloudminds Shenzhen Robotics Systems Co Ltd
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 Cloudminds Shenzhen Robotics Systems Co Ltd filed Critical Cloudminds Shenzhen Robotics Systems Co Ltd
Priority to PCT/CN2018/086876 priority Critical patent/WO2019218152A1/zh
Priority to CN201880001303.1A priority patent/CN108885814B/zh
Publication of WO2019218152A1 publication Critical patent/WO2019218152A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07F—COIN-FREED OR LIKE APPARATUS
    • G07F9/00—Details other than those peculiar to special kinds or types of apparatus
    • G07F9/002—Vending machines being part of a centrally controlled network of vending machines
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07F—COIN-FREED OR LIKE APPARATUS
    • G07F9/00—Details other than those peculiar to special kinds or types of apparatus
    • G07F9/02—Devices for alarm or indication, e.g. when empty; Advertising arrangements in coin-freed apparatus
    • G07F9/026—Devices for alarm or indication, e.g. when empty; Advertising arrangements in coin-freed apparatus for alarm, monitoring and auditing in vending machines or means for indication, e.g. when empty
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/08—Payment architectures
    • G06Q20/18—Payment architectures involving self-service terminals [SST], vending machines, kiosks or multimedia terminals
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07F—COIN-FREED OR LIKE APPARATUS
    • G07F9/00—Details other than those peculiar to special kinds or types of apparatus
    • G07F9/10—Casings or parts thereof, e.g. with means for heating or cooling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details

Definitions

  • the present application relates to the field of smart retailing, and in particular to a smart vending cabinet.
  • the operation mode of the intelligent vending cabinet is usually three processes of scanning the code to open the door, the user selects the item, and closes the door automatically, which is convenient and quick and does not need to be on duty.
  • the inventors found that the current smart vending cabinet usually installs a camera in the container, and the image is captured by the camera and visually identifies the type and quantity of the items in the container through artificial intelligence (AI). Since the identification of the type of goods and the amount of data in the container depends on the image captured by the camera, if the captured image is not clear or the image of the product in the captured image is not clear, the difficulty of AI recognition is greatly increased, and the accuracy of recognition is reduced. At the same time, the cost of using smart vending cabinets has also increased. Therefore, how to improve the image quality of the camera to collect all the goods in the intelligent vending cabinet is an urgent problem to be solved.
  • AI artificial intelligence
  • the technical problem to be solved in some embodiments of the present application is to provide an intelligent vending cabinet, which improves the quality of collecting product images in the intelligent vending cabinet, improves the accuracy of product identification, and reduces the use cost of the intelligent container.
  • An embodiment of the present application provides a smart vending cabinet, including: a cabinet, at least one layer partition, a camera, and a main control module, wherein the camera is communicatively coupled to the main control module; at least one layer partition is installed in the cabinet For separating the cabinet into M layer spaces, the middle of the layer partition is high and the height is gradually reduced to both ends, the layer partition is spanned between the opposite sides of the cabinet, and M is an integer greater than 1; A camera is arranged at the top of the space, and the camera captures the objects in the layer space of the camera under the control of the main control module; the main control module acquires the image captured by the camera and transmits the image to the image processing device for processing.
  • a camera is disposed at the top of each layer space, and the camera shooting angle is a plan view. Since the shape of each layer partition is middle height and the height gradually decreases toward both ends, There is a tilt angle between the placed object placed on the layer partition and the horizontal plane, which reduces the occlusion between adjacent objects, increases the area taken by the camera for the object, and increases the effective pixels of the object in the captured image. Improves the accuracy of the identification of the objects in the image; at the same time, because it identifies the items in each layer space, without identifying the items in the entire vending cabinet at one time, the accuracy of the item identification is improved, and the layer is enhanced. The accuracy of the identification of objects placed in the space.
  • FIG. 1 is a schematic perspective view of a smart vending machine in a first embodiment of the present application
  • FIG. 2 is a front view of the smart vending cabinet in the first embodiment of the present application.
  • FIG. 3 is a front view of the smart vending cabinet in the first embodiment of the present application.
  • FIG. 4 is a schematic view of a middle partition of a smart vending cabinet in a first embodiment of the present application
  • FIG. 5 is a schematic perspective structural view of a smart vending cabinet in a first embodiment of the present application.
  • FIG. 6 is a schematic view of a layer partition of a smart vending cabinet in a second embodiment of the present application.
  • FIG. 7 is a schematic view of a layer partition of a smart vending cabinet in a third embodiment of the present application.
  • Figure 8 is a schematic view showing a deposit on a layer partition of a smart vending cabinet in a third embodiment of the present application.
  • Figure 9 is a schematic view of the center line of the placed object in the third embodiment of the present application.
  • FIG. 10 is a schematic view of a center line of a main control module of a smart vending cabinet in a third embodiment of the present application.
  • the first embodiment of the present application relates to a smart vending cabinet that can be placed in a supermarket, subway, theater, etc. where no one is selling, so that people can shop on their own.
  • the smart vending cabinet 10 includes a cabinet 101, at least one layer partition 102, a camera 103 and a main control module 104.
  • the camera 103 is communicatively coupled to the main control module 104.
  • the specific structure is shown in Figure 1.
  • At least one layer partition 102 is installed in the cabinet 101 for dividing the cabinet 101 into M layer spaces, and the layer partition 102 is high in the middle and the height is gradually lowered toward both ends.
  • M is an integer greater than 1; the top of the layer space is provided with a camera 103, and the camera 103 captures the placement in the layer space of the camera under the control of the main control module 104; The main control module 104 acquires an image taken by the camera and transmits the image to the image processing device for processing.
  • two layers of partitions divide the cabinet 101 into three layer spaces.
  • the number of the layer partitions 102 can be determined according to the actual needs and the height of the cabinet 101, and the cabinet body 101 is divided into a plurality of layer spaces by the layer partition 102.
  • the height of the cabinet body is 1.5 meters, and the height of the objects in the actual life.
  • the specific shape of the cabinet 101 is not limited.
  • the cabinet 101 may be a rectangular parallelepiped or a trapezoidal shape. The specific shape may be determined according to actual needs.
  • the cabinet 101 is described by taking a rectangular parallelepiped as an example.
  • a plurality of mutually symmetric card slots may be respectively disposed between the opposite two sides.
  • three sides are respectively disposed on the two sides.
  • Three mutually symmetrical card slots, three pairs of card slots correspond to three types of layer spacers.
  • the shape of the layer baffle 102 is as shown in FIG. 1 , the middle is high, and the height is gradually lowered toward both ends.
  • the layer baffle 102 is an arc as an example, and it should be noted that the baffle plate is further Other shapes may be used, for example, the layer spacer 102 may also be a triangle as shown in FIG.
  • the main control module is not shown
  • a trapezoid as shown in FIG. 3 the trapezoid may be an isosceles trapezoid
  • the main control module is not shown in the drawings.
  • the application is not limited to the layer separators of several shapes listed in the embodiment, and may be other layer separators that satisfy the intermediate height and gradually decrease in height toward both ends.
  • each layer space is provided with a camera 103.
  • the camera 103 can be installed in the middle of the top of the current layer space, and the camera lens is facing the bottom of the current layer space. in the middle.
  • the camera 103 can be installed at other positions on the top of the layer space, which is not limited in this embodiment.
  • the height of the layer spacer 102 is determined according to the viewing angle of the camera disposed on the layer spacer and the width of the layer space captured by the camera disposed on the layer spacer.
  • the range of viewing angles of the cameras within each layer space can cover the entire layer space.
  • the viewing angle of the camera disposed on the layer spacer is determined and the width of the layer space photographed by the camera disposed on the layer spacer is determined, the height of the corresponding layer spacer 102 can be determined.
  • the layer spacer 102 is taken as an example of an arc to explain in detail how to determine the height of the layer spacer.
  • the curvature of the curved layer spacer can be determined by the angle of view of the camera.
  • the arc formula of the layer separator can be:
  • ⁇ is the curvature of the layer of the layer
  • ⁇ is the number of viewing angles of the camera.
  • the height of the curved layer partition is as shown in FIG. 4, assuming that the width of the layer space (as shown in FIG. 4) of the camera provided on the layer partition is 80 cm, the camera angle is 150 degrees, and due to the camera. Located at the center of the partition of the layer, the height of the curved layer partition is calculated to be equal to 8 cm (the value is rounded off) and the arc is 30 degrees by a trigonometric function in mathematics.
  • the layer partition is triangular or trapezoidal
  • the trigonometric function is adopted.
  • the height of the corresponding layer partition can be solved. For example, if the layer partition is triangular, the camera is located at the center of the layer partition and the viewing angle is 150 degrees, and the width of the layer space taken by the camera provided on the layer partition is 80. In centimeters, the height of the corresponding layer separator can be determined by the tangent formula in the trigonometric function to be 8 cm. The calculation of the height of the layer separator when the layer separator is trapezoidal will not be enumerated here.
  • the placed object is placed on the layer partition plate.
  • the surface of the layer partition facing the camera may be a rough surface, increasing. The friction between the layer partition and the placement. Of course, there are other ways to prevent the placement from slipping.
  • the camera 103 is mounted on the top of the layer space, and the lens faces the bottom of the cabinet, in a top view manner.
  • the layer partition 102 is high in the middle, and when the placed object is placed, a low-level placement can be placed in the middle position, and the height-high placement is placed at a position where the height of the two ends is low; since the layer partition 102 is high in the middle, The height gradually decreases from the middle to the both ends, so that there is a height difference between the objects, which increases the area photographed by the camera.
  • the smart vending cabinet 10 further includes a cabinet door 105, and the cabinet door 105 is hinged with any one of the opposite two sides; the main control module 104 controls the camera 103 to the camera 103 after detecting that the cabinet door 105 is closed. The objects in the layer space are taken for shooting.
  • the specific structure of the smart vending cabinet 10 is shown in FIG.
  • a wired connection may be used between the main control module 104 and the camera 103, such as a network cable or a USB data cable, or a wireless connection, such as a wireless fidelity (wifi) or a wireless communication network (3G/4G/5G). Either way; or use wired connection and wireless connection to ensure the data is not transmitted.
  • the detection of whether the cabinet door is closed by the main control module 104 can be realized by a sensor mounted on the cabinet door 105 or on either side of the two sides.
  • the main control module 104 is communicatively connected to the sensor.
  • the main control module 104 After the main control module 104 detects that the cabinet door is closed by the sensor, the main control module 104 controls the imaging 103 to capture the placement in the current layer space, and transmits the captured image to the image processing device. Image recognition is performed to determine the type of the placed object and the corresponding number in the captured image.
  • the detection method of the door of the main control module 104 is not limited to the method listed in the embodiment, and may be detected by other methods, which are not enumerated here.
  • a camera is disposed at the top of each layer space, and the camera shooting angle is a plan view. Since the shape of each layer partition is middle height and the height gradually decreases toward both ends, There is a tilt angle between the placed object placed on the layer partition and the horizontal plane, which reduces the occlusion between adjacent objects, increases the area taken by the camera for the object, and increases the effective pixels of the object in the captured image. Improves the accuracy of the identification of the objects in the image; at the same time, because it identifies the items in each layer space, without identifying the items in the entire vending cabinet at one time, the accuracy of the item identification is improved, and the layer is enhanced. The accuracy of the identification of objects placed in the space.
  • the second embodiment of the present application relates to a smart vending cabinet.
  • the smart vending cabinet in this embodiment is a further improvement of the smart vending container in the first embodiment.
  • the main improvement is that in the embodiment, the layer baffle At least one divider strip is provided on one side facing the top of the cabinet for securing the placement placed on the layer partition.
  • the specific structure of the layer separator is as shown in FIG. 6.
  • At least one partitioning strip 106 is disposed on a side of the layer partition 102 facing the top of the cabinet.
  • the number of the partitioning strips 106 can be set according to actual needs, for example, if the triangular layer partition faces the cabinet.
  • the top side of the body is smoother, and the two slopes of the triangular layer partition can hold up to 8 boxes of milk, so three dividers can be placed on each slope.
  • the partitioning strip 106 can be snapped onto the layer partition 102, and can be integrally formed with the layer partition 102 or other connection manner with the layer partition 102, which will not be enumerated in this embodiment.
  • the height of the partition strips may be smaller than the height of the objects.
  • the height of each of the partition strips may be different.
  • the height of the partition strips in this embodiment is as shown in FIG. 6.
  • the shape of the separator is not particularly limited in this embodiment, and may be a wave shape or a straight shape.
  • the smart vending cabinet provided in the embodiment provides at least one partition on the side of the layer partition facing the top of the cabinet, so as to facilitate the fixing of the placement on the layer partition, and at the same time, the layer is provided by the partition strip
  • the partitions are divided into rows to facilitate placement of the objects in columns.
  • the third embodiment of the present application relates to a smart vending cabinet.
  • the smart vending cabinet in this embodiment is a further improvement to the smart vending cabinet in the second embodiment.
  • the main improvement is that in this embodiment, there is a separation.
  • the strips are disposed at the center of the layer partition, and the partitions disposed on both sides of the center position are rotatable.
  • the specific structure of the layer separator is as shown in FIG.
  • the partition strips 106 disposed at the center of the layer partition 102 are perpendicular to the horizontal plane; each of the partition strips 106 disposed on both sides of the center position is rotated on the layer partition 102 centering on the current fixed point until photographing
  • the intersection line coincides with the center line of the placed object, wherein the intersection line is the intersection between the face of the camera's viewing angle range and the face of the object fixed by the divider.
  • the partitioning strip 106 provided at the center of the layer partition 102 is fixed and the partitioning strip cannot be rotated.
  • the partition strips 106 disposed on both sides of the center position are movably connected with the layer partitions, for example, the rotating partition strips 106 can be fixed by snaps.
  • the rotation of the partitioning strip 106 can be controlled by the main control module.
  • the partitioning strip and the layer partitioning plate are connected by a gear, and the rotation of the gear is controlled by the motor to control the rotation of the dividing strip.
  • the black dot is the fixed point of the divider.
  • a box of milk 20 is placed in the current layer space as shown in FIG. 8, and an image of each angle of the boxed milk and a midline information of the largest side of the pixel area in each image are stored in advance, for example, assumed to be photographed.
  • the milk is shown in Figure 9 (front view of the milk carton), and the center lines are 201 and 202.
  • the master module controls the divider strip 106 to rotate slowly and controls the camera to take an image of the boxed milk. Each time the image is taken, the main control module obtains the angle between the separation bar and the layer partition, and after obtaining the captured image, calculates the intersection information between the side of the image facing the camera and the face of the camera angle range.
  • the black solid dots in Fig. 8 indicate the fixed points between the separator and the layer separator.
  • the image information of the pre-stored placement may be stored in the main control module, or may be stored in an image processing device communicatively coupled to the main control module. Similarly, when the separation bar is rotated, the image information is taken.
  • the analysis performed by the image may be a main control module or an image processing device that is connected to the main control module. This embodiment is not limited, and may be set according to actual needs.
  • the height of the divider within the layer space is determined based on the height of the placement, wherein the height of the divider at the center location is less than the height of each divider disposed on either side of the center location.
  • the height of the partition strip at the center position is smaller than the height of each partition strip disposed on both sides of the center position.
  • the heights of the strips on both sides of the center position may also be different.
  • the height of the divider on both sides of the center position ranges from one third to one half of the height of the highest placement, and the height of the divider can effectively prevent the divider from being too high. Affect the take-up.
  • the smart vending cabinet provided in the embodiment can adjust the shooting area of the placed object by rotating the dividing strip to ensure the largest shooting area of the object, thereby improving the pixel area of the placed object in the captured image, improving the quality of the captured image, and improving the quality.
  • the accuracy of the image analysis, at the same time, the divider can also fix the placement to ensure that the placement does not slip.
  • the main control module 104 is configured as shown in FIG. 10, and includes at least one processor 1041; and a memory 1042 communicably coupled to the at least one processor 1041.
  • the memory 1042 stores instructions that are executable by at least one processor 1041.
  • the specific structure of the main control module 104 is as shown in FIG.
  • Memory 1042 and processor 1041 are connected in a bus manner, and the bus can include any number of interconnected buses and bridges that link together one or more processors 1041 and various circuits of memory 1042.
  • the bus can also link various other circuits, such as peripherals, voltage regulators, and power management circuits, as is well known in the art and, therefore, will not be further described herein.
  • the bus interface provides an interface between the bus and the transceiver.
  • the transceiver can be an element or a plurality of elements, such as multiple receivers and transmitters, providing means for communicating with various other devices on a transmission medium.
  • the data processed by the processor 1041 is transmitted over the wireless medium via an antenna. Further, the antenna also receives the data and transmits the data to the processor 1041.
  • the processor 1041 is responsible for managing the bus and normal processing, and can also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions.
  • the memory 1042 can be used to store data used by the processor when performing operations.

Landscapes

  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Signal Processing (AREA)
  • Cash Registers Or Receiving Machines (AREA)
  • Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)

Abstract

本申请涉及智能零售领域，尤其涉及一种智能售货柜（10）。本申请提供了一种智能售货柜（10），包括：柜体（101）、至少一个层隔板（102）、摄像头（103）和主控模块（104），其中，摄像头（103）与主控模块（104）通信连接；至少一个层隔板（102）安装于柜体（101）内，用于将柜体（101）分隔为M个层空间，层隔板（102）中间高且高度逐渐向两端降低，层隔板（102）跨接于柜体（101）相对的两个侧面之间，M为大于1的整数；层空间的顶部设置有摄像头（103），摄像头（103）在主控模块（104）的控制下拍摄摄像头（103）所在层空间内的放置物；主控模块（104）获取摄像头（103）拍摄的图像，并将图像传输至图像处理装置进行处理。该智能售货柜（10），提高了采集智能售货柜（10）内商品图像的质量，提高商品识别的准确性，同时降低智能货柜的使用成本。

Description

一种智能售货柜 技术领域
本申请涉及智能零售领域，尤其涉及一种智能售货柜。
 背景技术
随着科技的不断发展，为了便于人们随时购买物品，出现了智能售货柜。智能售货柜的运行模式通常是扫码打开柜门、用户选用物品、关门自动结算三个过程，方便快捷且无需人值守。
 技术问题
发明人在研究现有技术过程中发现，目前的智能售货柜通常在货柜中安装摄像头，由摄像头采集图像并通过人工智能（AI）视觉识别货柜内的物品种类和数量。由于对货柜内商品种类和数据量的识别依赖于摄像头采集的图像，若采集的图像不清晰或采集到的图像中商品的图像不清晰，将大大增加AI识别的困难，降低识别的准确性，同时还增加了智能售货柜的使用成本。因而，如何提高摄像头采集智能售货柜内所有商品的图像质量是亟待解决的问题。
 技术解决方案
本申请部分实施例所要解决的技术问题在于提供一种智能售货柜，提高了采集智能售货柜内商品图像的质量，提高商品识别的准确性，同时降低智能货柜的使用成本。
本申请的一个实施例提供了一种智能售货柜，包括：柜体、至少一个层隔板、摄像头和主控模块，其中，摄像头与主控模块通信连接；至少一个层隔板安装于柜体内，用于将柜体分隔为M个层空间，层隔板中间高且高度逐渐向两端降低，层隔板跨接于柜体相对的两个侧面之间，M为大于1的整数；层空间的顶部设置有摄像头，摄像头在主控模块的控制下拍摄摄像头所在层空间内的放置物；主控模块获取摄像头拍摄的图像，并将图像传输至图像处理装置进行处理。
 有益效果
相对于现有技术而言，本申请部分实施例中每个层空间的顶部设置有摄像头，且摄像头拍摄角度为俯视，由于每个层隔板的形状为中间高且高度逐渐向两端降低，使得放置在该层隔板上的放置物与水平面之间存在倾斜角，减少了相邻放置物之间的遮挡，增加了摄像头对放置物拍摄的面积，增加拍摄图像中放置物的有效像素，提高了对图像中放置物识别的准确性；同时由于是对每一个层空间内物品进行识别，而无需一次识别整个售货柜内的物品，提高了对物品识别的准确性，增强对每个层空间内放置物的识别准确性。
 附图说明
一个或多个实施例通过与之对应的附图中的图片进行示例性说明，这些示例性说明并不构成对实施例的限定，附图中具有相同参考数字标号的元件表示为类似的元件，除非有特别申明，附图中的图不构成比例限制。
图1是本申请第一实施例中智能售货柜的立体结构示意图；
图2是本申请第一实施例中智能售货柜正面示意图；
图3是本申请第一实施例中智能售货柜正面示意图；
图4是本申请第一实施例中智能售货柜中层隔板的示意图；
图5是本申请第一实施例中智能售货柜的立体结构示意图；
图6是本申请第二实施例中智能售货柜的层隔板的示意图；
图7是本申请第三实施例中智能售货柜的层隔板的示意图；
图8是本申请第三实施例中智能售货柜的层隔板上放置物的示意图；
图9是本申请第三实施例中放置物的中线示意图；
图10是本申请第三实施例中智能售货柜的主控模块的中线示意图。
 本发明的实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请部分实施例进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。然而，本领域的普通技术人员可以理解，在本申请的各实施例中，为了使读者更好地理解本申请而提出了许多技术细节。但是，即使没有这些技术细节和基于以下各实施例的种种变化和修改，也可以实现本申请所要求保护的技术方案。
本申请的第一实施例涉及一种智能售货柜，该智能售货柜可放置在无人售货的超市、地铁、影院等地方，以便人们自行购物。该智能售货柜10包括：柜体101、至少一个层隔板102、摄像头103和主控模块104，摄像头103与主控模块104通信连接。具体的结构如图1所示。
一个具体的实现中，至少一个层隔板102安装于柜体101内，用于将柜体101分隔为M个层空间，层隔板102中间高且高度逐渐向两端降低，层隔板102跨接于柜体101相对的两个侧面之间，M为大于1的整数；层空间的顶部设置有摄像头103，摄像头103在主控模块104的控制下拍摄摄像头所在层空间内的放置物；主控模块104获取摄像头拍摄的图像，并将图像传输至图像处理装置进行处理。
具体的说，如图1所示，2个层隔板将柜体101分隔为3个层空间。层隔板102的数量可以根据实际需要和柜体101的高度确定，通过层隔板102将柜体101分成多个层空间，例如，柜体的高度为1.5米，实际生活中放置物的高度为170mm，那么为了便于拿出该放置物，假设每个层空间中最高的高度为200mm，那么可以设置7个层隔板。当然，本实施例中不限制柜体101的具体形状，柜体101可以长方体、也可以是梯形等，具体的形状可以根据实际需要确定，本实施例中柜体101以长方体为例进行说明。
在柜体101相对的两个侧面上分被设置有相互对称的卡槽（图1中未示出），层隔板102通过对称的卡槽固定于相对的两个侧面之间。
可以理解的是，为了便于调节层空间的大小，可以在相对的两个侧面之间分别设置多个相互对称的卡槽，例如，假设有1个层隔板，那么在两个侧面分别设置3个相互对称的卡槽，3对卡槽对应3种层隔板放置方式。层隔板102的形状如图1中所示，中间高，且高度逐渐向两端降低，其中，图1中是以层隔板102为弧形为例，需要说明的是，层隔板还可以是其他形状，例如层隔板102还可以是如图2（图中未示出主控模块）中所示的三角形，或者，如图3中所示的梯形（梯形可以是等腰梯形，图中未示出主控模块），当然，应用中不限于本实施例中列举的几种形状的层隔板，还可以是其他满足中间高且高度向两端逐渐降低的层隔板。
每个层空间的顶部设置有摄像头103，可以理解的是，为了便于摄像头对层空间进行拍摄，可以将摄像头103安装在当前层空间的顶部的中间位置，摄像头的镜头正对当前层空间的底部中间位置。当然，摄像头103可以安装在层空间的顶部的其他位置，本实施例中不做限制。
一个具体的实现中，层隔板102的高度是根据层隔板上设置的摄像头的视角和层隔板上设置的摄像头所拍摄的层空间的宽度确定。
具体的说，为了确保对每个层空间的图像拍摄的准确以及后续对该图像识别的准确性，应当确保每个层空间内的摄像头的视角范围可以覆盖整个层空间。当确定了设置在层隔板上的摄像头的视角以及确定了层隔板上设置的摄像头所拍摄的层空间的宽度，即可确定出对应的层隔板102的高度。下面将以层隔板102为弧形为例，详细说明如何确定层隔板的高。
当层隔板为弧形，如图4所示，弧形层隔板的弧度可以通过摄像头的视角确定。该层隔板的弧度公式可以为：
α=(180°-β)÷360×2π，
其中，α为该层隔板的弧度，β为摄像头的视角度数。而弧形层隔板的高如图4所示，假设层隔板上设置的摄像头所拍摄的层空间的宽度（如图4中的L）为80厘米，摄像头视角为150度，且由于摄像头位于该层隔板的中心位置，通过数学中的三角函数即可计算出弧形层隔板的高等于8厘米（该值为四舍五入后的值）、弧度为30度。
可以理解的是，如图2和图3所示，若层隔板为三角形或者为梯形，在确定层隔板上设置的摄像头所拍摄的层空间的宽度以及摄像头的视角后，通过三角函数即可求解出对应的层隔板的高，例如，假设层隔板为三角形，摄像头位于层隔板的中心位置且视角为150度，层隔板上设置的摄像头所拍摄的层空间的宽度为80厘米，通过三角函数中的正切公式即可求出对应的层隔板的高为8厘米。此处将不再列举层隔板为梯形时该层隔板高的计算。
此外，需要说明的是，安装好层隔板后，将放置物放置在层隔板上，为了防止放置物下滑，影响摄像头的拍摄，那么层隔板面向摄像头的表面可以为粗糙表面，增大层隔板与放置物之间的摩擦力。当然，还可以采用其他方式防止放置物下滑。
摄像头103安装在层空间的顶部，且镜头朝向柜体底部，采用俯视拍摄的方式。层隔板102中间高，在放置放置物时，在中间位置可以放置高度低的放置物，将高度高的放置物放置在中间位置两端高度低的位置；由于层隔板102中间高，且从中间向两端高度逐渐降低，使得放置物之间存在高度差，增加了摄像头拍摄的面积。
一个具体的实现中，智能售货柜10还包括柜门105，柜门105与相对的两个侧面中的任意一个铰接；主控模块104在检测到柜门105关闭后，控制摄像头103对摄像头103所在层空间内的放置物进行拍摄。智能售货柜10的具体结构如图5所示。
具体的说，主控模块104与摄像头103之间可以采用有线连接，如，网线、USB数据线等；或者采用无线连接，如无线保真（wifi）、无线通信网络（3G/4G/5G）等方式；或者即采用有线连接又采用无线连接的方式，确保数据的无障碍传输。主控模块104对柜门是否关闭的检测，可以通过柜门105上或者两个侧面中任一面安装的传感器来实现。主控模块104与传感器通信连接，主控模块104通过传感器检测到柜门关闭后，主控模块104控制摄像103对当前层空间内的放置物进行拍摄，并将拍摄的图像传输至图像处理装置进行图像识别，确定拍摄图像中的放置物种类和对应的数量。
可以理解的是，主控模块104对柜门的检测方法并不限制于本实施例中列举的方法，还可以采用其他方式进行检测，此处不再一一进行列举。
相对于现有技术而言，本申请部分实施例中每个层空间的顶部设置有摄像头，且摄像头拍摄角度为俯视，由于每个层隔板的形状为中间高且高度逐渐向两端降低，使得放置在该层隔板上的放置物与水平面之间存在倾斜角，减少了相邻放置物之间的遮挡，增加了摄像头对放置物拍摄的面积，增加拍摄图像中放置物的有效像素，提高了对图像中放置物识别的准确性；同时由于是对每一个层空间内物品进行识别，而无需一次识别整个售货柜内的物品，提高了对物品识别的准确性，增强对每个层空间内放置物的识别准确性。
本申请的第二实施例涉及一种智能售货柜，本实施例中的智能售货柜是对第一实施例中智能售货柜的进一步改进，主要改进之处在于，本实施例中，层隔板朝向柜体顶部的一面上设置有至少一个分隔条，该分隔条用于固定放置于层隔板上的放置物。该层隔板的具体结构如图6所示。
具体的说，如图6所示，在层隔板102朝向柜体顶部的一面上设置至少一个分隔条106，分隔条106的数目可以根据实际需要设定，例如，若三角形层隔板朝向柜体顶部的一面上较为光滑，且三角形层隔板的两个斜面最多可以放置8盒牛奶，那么可以在每个斜面上设置3个分隔条。分隔条106可以卡接在层隔板102上，可以是与层隔板102一体成型，或者与层隔板102采用其他的连接方式，本实施例中不再一一列举。
可以理解的是，分隔条的高度可以小于放置物的高度，当然，每一个分隔条的高度都可以不一样，本实施例中的分隔条的高度如图6中所示。分隔条的形状本实施例中不做具体限制，可以为波浪形，或者直线形状。
本实施例中提供的智能售货柜，通过在层隔板朝向柜体顶部的一面上设置至少一个分隔条，便于固定放置在层隔板上的放置物，同时，由于设置了分隔条，将层隔板分成多列，便于放置物按列放置。
本申请的第三实施例涉及一种智能售货柜，本实施例中的智能售货柜是对第二实施例中智能售货柜的进一步改进，主要改进之处在于，本实施例中，有一个分隔条设置在层隔板的中心位置，设置在中心位置两侧的分隔条可以转动。该层隔板的具体结构如图7所示。
一个具体实现中，在层隔板102的中心位置设置的分隔条106垂直于水平面；设置于中心位置两侧的每一个分隔条106在层隔板102上以当前固定点为中心转动，直至拍摄交线与放置物的中线重合，其中，拍摄交线为摄像头的视角范围所在面与由分隔条固定的放置物的被拍摄面之间的交线。
具体的说，如图7所示，在层隔板102的中心位置设置的分隔条106固定且该分隔条不能转动。而设置在中心位置两侧的分隔条106与层隔板之间采用活动连接，例如，可以通过卡扣固定转动的分隔条106。可以理解的是，分隔条106的转动可以通过主控模块控制，例如，分隔条与层隔板之间通过齿轮连接，通过控制电机转动齿轮，从而控制分隔条的转动。图7中，黑色圆点为分隔条的固定点。
下面将详细介绍调整分隔条直至拍摄交线与放置物的中线重合的过程。
具体的说，假设在当前层空间中如图8所示放置有一盒牛奶20，预先存储该盒装牛奶的各个角度的图像以及每个图像中像素面积最大一面的中线信息，例如，假设拍摄的牛奶如图9（牛奶盒的正面视图），则中线为201和202。主控模块控制分隔条106缓慢转动，并控制摄像头拍摄该盒装牛奶的图像。在每次拍摄图像时，主控模块获取分隔条与层隔板之间的角度，在获得拍摄的图像后计算该图像中牛奶面向摄像头的一面与摄像头视角范围所在面之间的交线信息，并与预存的该盒装牛奶的中线信息进行比对，若确定拍摄交线与放置物的中线重合，则停止转动分隔条，否则继续转动。图8中黑色实心圆点表示分隔条与层隔板之间的固定点。
此外，需要说明的是，预先存储的放置物的图像信息可以存储在主控模块中，也可以是存储在与主控模块通信连接的图像处理装置中，同理，对分隔条转动时拍摄的图像进行的分析可以是主控模块，也可以是与主控模块通信连接的图像处理装置，本实施例对此不做限定，具体可以根据实际需要设定。
一个具体实现中，层空间内的分隔条的高度是根据放置物高度确定，其中，位于中心位置的分隔条的高度小于设置于中心位置两侧的每个分隔条的高度。
具体的说，如图8示，位于中心位置的分隔条的高度小于设置于中心位置两侧的每个分隔条高度，当然，中心位置两侧的分隔条高度也可以都不相同。一个具体实施方案中，中心位置两侧的分隔条高度的取值范围为最高放置物高度的三分之一至二分之一之间，在该高度范围内，可以有效避免分隔条太高而影响拿取放置物。
本实施例中提供的智能售货柜，通过转动分隔条，从而可以调整放置物的拍摄面积，确保放置物的拍摄面积最大，从而提高拍摄图像中放置物像素面积，提高拍摄图像的质量，提高对图像分析的准确性，同时，分隔条还可以固定放置物，确保放置物不会滑动。
此外，需要说明的是，上述实施例中，主控模块104的结构如图10所示，包括至少一个处理器1041；以及，与至少一个处理器1041通信连接的存储器1042。存储器1042存储有可被至少一个处理器1041执行的指令。主控模块104的具体结构如图10所示。
存储器1042和处理器1041采用总线方式连接，总线可以包括任意数量的互联的总线和桥，总线将一个或多个处理器1041和存储器1042的各种电路链接在一起。总线还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起，这些都是本领域所公知的，因此，本文不再对其进行进一步描述。总线接口在总线和收发机之间提供接口。收发机可以是一个元件，也可以是多个元件，比如多个接收器和发送器，提供用于在传输介质上与各种其他装置通信的单元。经处理器1041处理的数据通过天线在无线介质上进行传输，进一步，天线还接收数据并将数据传送给处理器1041。
处理器1041负责管理总线和通常的处理，还可以提供各种功能，包括定时，外围接口，电压调节、电源管理以及其他控制功能。而存储器1042可以被用于存储处理器在执行操作时所使用的数据。
本领域的普通技术人员可以理解，上述各实施例是实现本申请的具体实施例，而在实际应用中，可以在形式上和细节上对其作各种改变，而不偏离本申请的精神和范围。

Claims (11)

  1. 一种智能售货柜，其中，包括：柜体、至少一个层隔板、摄像头和主控模块，其中，所述摄像头与所述主控模块通信连接；
    所述至少一个层隔板安装于所述柜体内，用于将所述柜体分隔为M个层空间，所述层隔板中间高且高度逐渐向两端降低，所述层隔板跨接于所述柜体相对的两个侧面之间，M为大于1的整数；
    所述层空间的顶部设置有所述摄像头，所述摄像头在所述主控模块的控制下拍摄所述摄像头所在层空间内的放置物；
    所述主控模块获取所述摄像头拍摄的图像，并将所述图像传输至图像处理装置进行处理。
  2. 根据权利要求1所述的智能售货柜，其中，所述层隔板的高度是根据所述层隔板上设置的摄像头的视角和所述层隔板上设置的摄像头所拍摄的层空间的宽度确定。
  3. 根据权利要求1或2所述的智能售货柜，其中，所述层隔板的形状为弧形、三角形或梯形。
  4. 根据权利要求1至3中任一项所述的智能售货柜，其中，所述层隔板朝向所述柜体顶部的一面上设置有至少一个分隔条，所述分隔条用于固定放置于所述层隔板上的放置物。
  5. 根据权利要求4所述的智能售货柜，其中，有一个所述分隔条设置在所述层隔板的中心位置。
  6. 根据权利要求5所述的智能售货柜，其中，在所述层隔板的中心位置设置的分隔条垂直于水平面；
    设置于所述中心位置两侧的每个所述分隔条在所述层隔板上以当前固定点为中心转动，直至拍摄交线与所述放置物的中线重合，其中，所述拍摄交线为所述摄像头的视角范围所在面与由所述分隔条固定的所述放置物的被拍摄面之间的交线。
  7. 根据权利要求5或6所述的智能售货柜，其中，所述层空间内的所述分隔条的高度是根据放置物高度确定，其中，位于所述中心位置的所述分隔条的高度小于设置于所述中心位置两侧的每个所述分隔条的高度。
  8. 根据权利要求7所述的智能售货柜，其中，所述设置于所述中心位置两侧的分隔条高度的取值范围为：最高放置物高度的三分之一至二分之一之间。
  9. 根据权利要求1至8中任一项所述的智能售货柜，其中，所述相对的两个侧面上分别设置有相互对称的卡槽，所述层隔板通过对称的卡槽固定于所述相对的两个侧面之间。
  10. 根据权利要求9所述的智能售货柜，其中，所述智能售货柜还包括柜门，所述柜门与所述相对的两个侧面中的任意一个铰接；
    所述主控模块在检测到柜门关闭后，控制所述摄像头对所述摄像头所在层空间内的放置物进行拍摄。
  11. 根据权利要求1至10中任一项所述的智能售货柜，其中，所述摄像头与所述主控模块采用有线和/或无线通信的方式连接。
PCT/CN2018/086876 2018-05-15 2018-05-15 一种智能售货柜 Ceased WO2019218152A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/086876 WO2019218152A1 (zh) 2018-05-15 2018-05-15 一种智能售货柜
CN201880001303.1A CN108885814B (zh) 2018-05-15 2018-05-15 一种智能售货柜

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/086876 WO2019218152A1 (zh) 2018-05-15 2018-05-15 一种智能售货柜

Publications (1)

Publication Number Publication Date
WO2019218152A1 true WO2019218152A1 (zh) 2019-11-21

Family

ID=64325012

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/086876 Ceased WO2019218152A1 (zh) 2018-05-15 2018-05-15 一种智能售货柜

Country Status (2)

Country Link
CN (1) CN108885814B (zh)
WO (1) WO2019218152A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112735030A (zh) * 2020-12-28 2021-04-30 深兰人工智能(深圳)有限公司 售货柜的视觉识别方法、装置、电子设备和可读存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109343124B (zh) * 2018-12-05 2020-11-27 六安正健信息技术有限公司 一种高铁储物柜闭合监测系统
CN109919111B (zh) * 2019-03-13 2023-07-04 创新先进技术有限公司 确定摄像头对应的货柜区域的方法、装置、货柜及系统
CN113408491B (zh) * 2021-07-20 2023-05-30 杭州海康机器人股份有限公司 物品获得结果的统计方法、装置、系统及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009155641A1 (en) * 2008-06-24 2009-12-30 Griffits John P Computer controlled object locating system
CN205942922U (zh) * 2016-07-28 2017-02-08 吉林工程技术师范学院 一种具有验视功能的物流货柜
CN107369265A (zh) * 2017-08-14 2017-11-21 武汉未远金易科技有限公司 一种组合集成式智能快递柜
CN107730760A (zh) * 2017-09-29 2018-02-23 合肥诺优信息科技有限公司 快递发收装置及方法
CN207164903U (zh) * 2017-06-30 2018-03-30 东莞市山河电脑科技有限公司 一种智能果蔬储存柜

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009155641A1 (en) * 2008-06-24 2009-12-30 Griffits John P Computer controlled object locating system
CN205942922U (zh) * 2016-07-28 2017-02-08 吉林工程技术师范学院 一种具有验视功能的物流货柜
CN207164903U (zh) * 2017-06-30 2018-03-30 东莞市山河电脑科技有限公司 一种智能果蔬储存柜
CN107369265A (zh) * 2017-08-14 2017-11-21 武汉未远金易科技有限公司 一种组合集成式智能快递柜
CN107730760A (zh) * 2017-09-29 2018-02-23 合肥诺优信息科技有限公司 快递发收装置及方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112735030A (zh) * 2020-12-28 2021-04-30 深兰人工智能(深圳)有限公司 售货柜的视觉识别方法、装置、电子设备和可读存储介质
CN112735030B (zh) * 2020-12-28 2022-08-19 深兰人工智能(深圳)有限公司 售货柜的视觉识别方法、装置、电子设备和可读存储介质

Also Published As

Publication number Publication date
CN108885814B (zh) 2019-11-22
CN108885814A (zh) 2018-11-23

Similar Documents

Publication Publication Date Title
CN110909611B (zh) 一种关注区域检测方法、装置、可读存储介质及终端设备
JP7351861B2 (ja) ユーザ識別子を指定するシステム及びその方法
US12333491B1 (en) Determining inventory changes at an inventory location
WO2019218152A1 (zh) 一种智能售货柜
US10909667B1 (en) Image rectification using transformation data
US12524460B2 (en) System and method for selecting an item from a plurality of identified items based on a similarity value
US9194931B2 (en) Length measurement method and device of the same
WO2019232703A1 (zh) 智能售货柜、物品识别的方法、装置、服务器和存储介质
EP2843590B1 (en) System and method for package dimensioning
US10169677B1 (en) Counting stacked inventory using image analysis
CN108780505A (zh) 智能售货柜、物品识别方法、装置、服务器和存储介质
JP2022043067A (ja) 対象物位置決めシステム
CN113850657B (zh) 基于无人车的商品售卖方法、装置、电子设备及存储介质
US20240029276A1 (en) System and method for identifying moved items on a platform during item identification
US12579783B2 (en) System and method for selecting an item from a plurality of identified items by filtering out back images of the items
US20170178333A1 (en) Methods and systems for occlusion detection and data correction for container-fullness estimation
CN115223514A (zh) 一种智能调节参数的液晶显示驱动系统及方法
CN106254839A (zh) 集装箱卡车防吊起方法及装置
CN107749981A (zh) 摄像头视场角测量方法、设备及系统
CN109002804A (zh) 人脸识别方法、装置、存储介质及电子设备
CN110532840A (zh) 一种方形物体的形变识别方法、装置及设备
JP6255819B2 (ja) 計測用コンピュータプログラム、計測装置及び計測方法
CN207677915U (zh) 摄像头视场角测量系统及设备
CN113674336B (zh) 货架空置信息确定方法、计算机设备及存储装置
US12567048B2 (en) Hardware system for identifying grab- and-go transactions in a cashierless store

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: 18919129

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 13/04/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18919129

Country of ref document: EP

Kind code of ref document: A1