FORM IDENTIFICATION DEVICE AND DELIVERY BOX

Abstract

A delivery box (12) comprising a shape identification device (20) and an operation display unit (17), wherein the shape identification device (20) comprises: - a receiving chamber (14, 21) for receiving a measurement target object (60) so that the measurement target object (60) can be placed therein and taken out, wherein the measurement target object (60) is a delivery package (60); - a first imaging unit (23) for imaging the measurement target object (60) received in the receiving chamber (14, 21); and - an image processing unit (30) for performing image processing based on information of an image captured by the first imaging unit (23) to recognize an external shape of the measurement target object (60), the image processing unit (30) including: ◯ a storage unit (35) for storing in advance a plurality of types of pattern outer shapes corresponding to the measurement target object (60), ◯ a shape recognition unit (32) for comparing outer shape information from the first imaging unit (23) with outer shape information stored in the storage unit (35) to perform identification, and for recognizing an outer shape that matches or approximates the outer shape information from the first imaging unit (23) as one of a rectangular parallelepiped shape, a cylindrical shape, and an elongated shape with a predetermined length or more that is different from the rectangular parallelepiped shape and the cylindrical shape, and o a dimension measuring unit (33) for measuring a dimension of the measurement target object (60) based on an external shape recognized by the shape recognition unit (32); wherein, when the shape recognition unit (32) recognizes the outer shape as the rectangular parallelepiped shape or the cylindrical shape, the dimension measuring unit (33) is configured to calculate a volume based on the measured dimension, and the delivery box (12) is configured to display a delivery fee corresponding to the calculation result on the operation display unit (17); and wherein, when the shape recognition unit (32) recognizes the outer shape as the elongated shape having the predetermined length or more, the delivery box (12) is configured to display a sending fee corresponding to a specific service on the operation display unit (17).

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a delivery box. BACKGROUND

[0002] As described in JP 2017-116482 A, a delivery box has been conventionally known that is installed in a public place where it can be used by unspecified persons and that includes a plurality of receiving chambers (bins). This delivery box can be used by a large number of unspecified persons to send delivery packages. When a sender places a delivery package into a bin of a measuring device of the delivery box, the dimension of the delivery package is measured by the measuring device. As a result, an optimal container for the delivery package is selected from containers of two sizes for delivery packages, and the door of the selected container is automatically opened. The sender places the delivery package into the selected container for the delivery package.

[0003] WO 2017 / 110071 A1 describes a size measuring device comprising: a housing chamber; an imaging unit that captures an image of the object to be measured accommodated in the housing chamber; and an image processing unit that processes the image captured by the imaging unit. The housing chamber consists of a rectangular bottom surface, a rectangular first side surface, and a rectangular second side surface. The housing chamber includes a first inner corner formed by the bottom surface and the first side surface, a second inner corner formed by the bottom surface and the second side surface, and a third inner corner formed by the first side surface and the second side surface. The housing chamber includes a reference corner formed at the intersection point of the first, second, and third inner corners.The housing chamber includes markings indicating the positions of the sides of the first side surface part, the second side surface part and the lower surface part.

[0004] JP 2017-187822 A describes a measuring system comprising: first measuring means for recording a weight determined by measuring a container and an object stored therein; second measuring means for recording a weight determined by measuring the container; imaging means for recording an image obtained by imaging the container with respect to the measurement by an imaging device during measurement by the first or second measuring means; identification means for identifying the container included in the image according to a container feature amount with respect to the image captured by the imaging means; specification means for specifying the weight of the container identified by the identification means with reference to information previously registered in a storage part;a calculation means for calculating the weight of the object by subtracting the weight of the container specified by the specification means from the weight of the container and the object measured by the first measuring means; and a registration means for registering the weight of the container measured by the second measuring means and the container characteristic quantity about the container in the identification means as information about the container in the storage part. SUMMARYTECHNICAL PROBLEM

[0005] The measuring device described in JP 2017-116482 A can automatically measure dimensions of measurement targets each having a rectangular parallelepiped shape or the like, where three surfaces to be imaged by an imaging unit are rectangular. Since such a measuring device limits the types of delivery package shapes that can be measured by the measuring device, there is room for improvement in expanding the types of measurement targets whose dimensions can be automatically measured. Furthermore, for automatically measuring the sizes of delivery packages, it is effective to automatically identify the shapes of the delivery packages.

[0006] An object of the disclosure is to provide a delivery box that can automatically identify the shape of a measurement target object accommodated in a receiving chamber. SOLUTION TO THE PROBLEM

[0007] The present invention relates to a delivery box according to claim 1. Claims 2 to 5 describe particularly advantageous realizations of the delivery box according to claim 1.

[0008] A shape identification device according to one aspect of the present disclosure includes: a receiving chamber for receiving a measurement target object so that the measurement target object can be placed therein and taken out; a first imaging unit for imaging the measurement target object received in the receiving chamber;and an image processing unit for performing image processing based on information of an image captured by the first imaging unit to recognize an external shape of the measurement target object, wherein the image processing unit includes a storage unit for storing a plurality of types of pattern external shapes corresponding to the measurement target object in advance and a shape recognition unit for comparing external shape information from the first imaging unit with external shape information stored in the storage unit to perform identification, and for recognizing an external shape that matches or approximates the external shape information from the first imaging unit.

[0009] A delivery box according to the present disclosure includes: the shape identifying device of the present disclosure; and a dimension measuring unit for measuring a dimension of the measurement target object based on an external shape recognized by the shape recognizing unit. ADVANTAGEOUS EFFECT OF THE INVENTION

[0010] By using the shape identification device and the delivery box of the present disclosure, it is possible to automatically identify the shape of a measurement target object accommodated in a receiving chamber. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a perspective view of a delivery box according to an embodiment of the present disclosure. Fig. 2 is a diagram showing a configuration of a Fig. 1 and a state in which a delivery request is communicated from the delivery box to a data subject. Fig. 3 is a perspective view showing a measuring device as a shape identifying device in the Fig. 1 shows the delivery box shown. Fig. 4 is a configuration diagram of the Fig. 3 shown measuring device. Fig. 5 is a flowchart showing shape recognition and dimension measurement in the embodiment. Fig. 6 is a flowchart showing shape recognition and dimension measurement in the embodiment in more detail. Fig. 7 is a perspective view showing an example of a receiving chamber of the measuring device. Fig. 8 is a diagram showing an edge image generated based on an image captured by an imaging unit when a delivery package as a measurement target object is a rectangular parallelepiped object. Fig. 9 is a diagram showing an edge image generated based on an image captured by the imaging unit when the delivery package has a cylindrical shape. Fig. 10 is a perspective view showing a measuring device as a shape identifying device according to another example of the embodiment. DESCRIPTION OF THE EMBODIMENT

[0011] An embodiment of the present disclosure will be described below with reference to the drawings. The shapes, numerical values, and numbers described below are examples for explanation and can be appropriately changed according to the specifications of the shape identification device and the delivery box. Hereinafter, similar elements are represented by the same reference numerals throughout the drawings. Furthermore, in the description, the previously described reference numerals will be used in the text where necessary.

[0012] In the following description and drawings, an X direction is a width direction of a delivery box 12, a Y direction is a depth direction of the delivery box 12, and a Z direction is a height direction of the delivery box 12. The X direction, the Y direction, and the Z direction are perpendicular to each other.

[0013] Fig. 1 is a perspective view of the delivery box 12 according to an embodiment of the present disclosure. Fig. 2 is a view showing a configuration of the delivery box 12 and a state in which a delivery request is notified from the delivery box 12 to a data subject. The delivery box 12 is installed in a public place where unspecified persons can use it, and can be used for sending delivery packages to a large number of unspecified persons.

[0014] As in Fig. 1, the delivery box 12 includes a substantially rectangular parallelepiped outer body (main body) 13, a plurality of delivery receiving chambers 14, doors 15 for opening and closing the openings of the respective delivery receiving chambers 14, a measuring device 20, and an operation display unit 17. The measuring device 20 corresponds to a shape identification device. The delivery receiving chamber 14 has a box-like shape for receiving a delivery package 60 as a measurement target object ( Fig. 3) so that it can be put in and taken out.

[0015] The door 15 blocks an opening towards a side in the depth direction Y (towards a front side of the paper surface of Fig. 1) of the delivery receiving chamber 14. The door 15 is a single swing-type door attached to the delivery box 12 and can swing about an axis (not shown) in the Z direction to open and close the opening of the delivery receiving chamber 14. The door 15 of the delivery receiving chamber 14 is locked by an electric lock 18 ( Fig. 2). The electric lock 18 is controlled by a user operation or the like on an operation display unit 17 ( Fig. 1). Note that the electric lock 18 can be configured to be unlocked by a barcode, a QR code (registered trademark), transmission and reception of a wireless signal using Bluetooth (registered trademark) with a portable terminal such as a smartphone, a contactless key using radio waves or infrared rays, an IC card, or input based on a numeric keypad.

[0016] A package sensor 19 ( Fig. 2) is provided inside the delivery receiving chamber 14. The package sensor 19 detects the presence or absence of a package, such as a delivery package, in the corresponding delivery receiving chamber 14. The package sensor 19 is formed, for example, by a strain gauge. The strain gauge includes a metallic resistance material, and the metallic resistance material contracts when a compressive force is applied thereto from the outside, and the resistance value thereof decreases. Detection information of the package sensor 19 is sent to a control device 30 ( Fig. 4) of the delivery box 12. The control device 30 detects the presence or absence of a parcel based on the weight of the parcel located in the delivery receiving chamber 14 based on the electrical signal. When a parcel is detected, the control device 30 converts the detection information into parcel detection information. The information detected by the parcel sensor 19 may be transmitted to a parcel sensor control unit (not shown) instead of the control device 30. In this case, the parcel sensor control unit converts the detection information of the parcel sensor 19 into parcel detection information. The parcel sensor 19 is provided in the delivery receiving chamber 14 and may be configured to detect the presence or absence of a parcel in the delivery receiving chamber 14 using an optical sensor, for example, an infrared sensor.

[0017] The measuring device 20 includes a measuring receiving chamber 21 which is formed at a lower end portion in the height direction Z of an end portion (a left end portion in Fig. 1) in the width direction X of the delivery box 12, and a door having an opening at one end (a front end of the paper surface of Fig. 1) in the depth direction Y of the measurement receiving chamber 21, allowing the opening to be opened and closed. The configuration and function of the door 22 are similar to those of the door 15 of the delivery receiving chamber 14. The door 22 of the measurement receiving chamber 21 can be configured so that it is not locked. The configuration of the measuring device 20 will be described in detail later.

[0018] The operation display unit 17 is arranged above the measuring device 20 on a side surface (the front side of the paper surface of Fig. 1) in the depth direction Y of the delivery box 12. The operation display unit 17 is constituted by a touch panel display or the like and has a function of accepting an input by a user operation and a function of a display unit. A settlement unit 70 is arranged below the operation display unit 17. The settlement unit 70 sends and receives wireless signals to, for example, an IC card possessed by a user, to settle accounts with electronic money. Note that the operation display unit 17 can be replaced by a configuration including a display unit constituted by an LCD display and an operation unit constituted by a numeric keypad or the like. The settlement unit 70 can be configured to settle accounts by inputting an identification number and password of a credit card via a numeric keypad.

[0019] When a sender of a delivery package who is a customer of a delivery company makes a delivery request for the delivery package using the delivery box 12, the delivery request is executed as follows. For example, when the sender selects a display section of "Send" from a menu screen displayed on the operation display unit 17 of the delivery box 12, the door 22 of the measuring device 20 automatically opens. When the sender places the delivery package in the measurement receiving chamber 21 and closes the door 22, the measuring device 20 detects the shape of the delivery package in the measurement receiving chamber 21 and then measures the size of the delivery package based on the shape. Further, the measuring device 20 calculates a volume as the size of the delivery package. The shape detected by the measuring device 20, the measured dimension, etc., are displayed on the operation display unit 17.

[0020] The sender uses the operation display unit 17 to input a delivery destination address, etc., and then takes out the delivery package from the metered receiving chamber 21. Note that the operation display unit 17 can be replaced with a configuration in which the delivery destination address, etc., can be registered in advance by pre-registration using the Internet. The sender then uses the operation display unit 17 to select one of the delivery receiving chambers 14. As a result, the door 15 of the selected delivery receiving chamber 14 is unlocked, and the door 15 is automatically opened. The sender therefore places the delivery package in the delivery receiving chamber 14. The package sensor 19 detects that the delivery package has been placed in the delivery receiving chamber 14. After that, a delivery fee is displayed on the operation display unit 17, and therefore the sender settles the delivery fee via the settlement unit 70.

[0021] When billing for the delivery package has been performed, information of a delivery request for the delivery package is transmitted from the delivery box 12 to a terminal 72 of a delivery company, a terminal 73 of the sender and a management server 74 via a communication network 71, as shown in Fig. 2. Therefore, the delivery company goes to an installation location of the delivery box 12, opens the door 15 of the corresponding delivery receiving chamber 14, takes out the delivery package from the delivery receiving chamber 14, and performs delivery to the sending destination.

[0022] Next, the measuring device 20 is described in detail. Fig. 3 is a perspective view showing the measuring device 20 in the delivery box 12. Fig. 4 is a configuration view of the measuring device 20. The measuring device 20 includes the above-described measurement receiving chamber 21 and the door 22, a first camera 23 for shape recognition, an illumination unit 24, and a control device 30. The first camera 23 corresponds to a first imaging unit, and the control device 30 corresponds to an image processing unit. The measuring device 20 is a shape identification device, and the control device 30 also has a dimension measuring unit 33 ( Fig. 4), which will be described later. In Fig. 3, the first camera 23, the illumination unit 24 and a distance sensor 40 described later are shown as if they are transparent.

[0023] The measurement receiving chamber 21 has a box-like shape for receiving a delivery package 60 so that it can be put in and taken out. As described above, the opening on one side in the depth direction Y (the front side on the paper surface in Fig. 3) of the measurement receiving chamber 21 is blocked by the door 22. The door 22 of the measurement receiving chamber 21 may be configured to be locked and unlocked with an electric lock, and the electric lock may be configured to be unlocked by a user operation or the like on the operation display unit 17 ( Fig. 1).

[0024] The first camera 23 is arranged to be directed obliquely downward to image the delivery package 60, for example, from a corner portion of an upper portion on an opening side of the measurement receiving chamber 21, for example, an end portion (a left end in Fig. 3) in the width direction X. For example, a CCD camera is used as the first camera 23. The first camera 23 may be a CMOS camera. The operation of the first camera 23 is controlled by the control device 30, and information about an image captured by the first camera 23 is transmitted to the control device 30. An image to be captured by the first camera 23 may be a grayscale image or a color image. If it is a color image, the image may be converted to a grayscale image during image processing. Note that Fig. 3 is a view in which the control device 30 is arranged on a cover plate 26 which forms an upper end of the measurement receiving chamber 21, but the present invention is not limited to this configuration, and the control device 30 may be arranged around the Fig. 1 shown operating display unit 17 of the delivery box 12.

[0025] The illumination unit 24 is arranged near the first camera 23 in the upper portion of the measurement receiving chamber 21 and illuminates the interior of the measurement receiving chamber 21, particularly a lower part thereof. The illumination unit 24 is controlled by the control device 30 to be turned on and off. For example, a white LED is used as the illumination unit 24. A fluorescent lamp, an incandescent bulb, or an LED of another color can be used as the illumination unit 24.

[0026] Note that the measuring device 20 may include a distance sensor 40, which is a measurement sensor. The distance sensor 40 is, for example, a distance image sensor. A distance image sensor includes, for example, a light source such as a near-infrared LED that flashes a light emission pulse at high speed and a CMOS image sensor, and measures, for each pixel, in real time, a time it takes for emitted light to strike an object and return. As a result, a distance image can be created that includes distance information indicating the longitude of the distance. For this reason, since the longitude of the surface of the object can also be detected from the acquired distance image, a three-dimensional external shape of the delivery package 60 can be more easily identified. The detection information of the distance sensor 40 is transmitted to the control device 30.Furthermore, the distance sensor 40 facilitates measuring the object's dimensions, such as length, width, and depth. The distance sensor 40 is used to measure the dimensions of the delivery package 60 accommodated in the measurement receiving chamber 21.

[0027] As in Fig. 4, the control device 30 includes, for example, an arithmetic processing unit 31 constituted by an MCU (Micro Controller Unit), and a storage unit 35. The storage unit 35 may be a RAM, ROM, or the like. The arithmetic processing unit 31 has a function of reading and executing a program or the like stored in advance in the storage unit 35. The storage unit 35 has a function of temporarily storing a readout program and processing data, and a function of pre-storing a control program, a predetermined threshold value, and the like. Further, images of a plurality of types of pattern outer shapes corresponding to outer shapes of delivery packages are stored in advance in the storage unit 35.

[0028] For example, various external shapes such as a rectangular parallelepiped, a column, a disk, and a hexagonal column can be considered for delivery packages. Generally, each type of external shape will have different images when viewed from different directions. For this reason, images captured from a plurality of different directions are preferably stored in advance in the storage unit 35 for each type of external shape.

[0029] When the control device 30 executes a program, the main function of the device, system, or method according to the present disclosure is implemented. The type of the arithmetic processing unit 31 is not limited as long as its function can be implemented by executing the program. For example, the arithmetic processing unit may be a CPU. The arithmetic processing unit may be constituted by one or more electronic circuits, including a semiconductor integrated circuit (IC) or an LSI (Large Scale Integration). The plurality of electronic circuits may be integrated on a single chip or provided on a plurality of chips. A plurality of chips may be integrated in a single device or provided in a plurality of devices.A non-volatile recording medium such as an optical disk or a hard disk drive can be used as the storage unit. An external storage device can be connected to the control device 30 as the storage unit. The program can be stored in advance on a recording medium or transferred to the recording medium via a wide area communication network including the Internet or the like.

[0030] The control device 30 also includes an interface 36. The operating display unit 17 ( Fig. 1) is connected to the arithmetic processing unit 31 via the interface 36. At least one of a personal computer (PC), an input device, and a monitor as a display may be connected to the arithmetic processing unit via the interface 36.

[0031] The arithmetic processing unit 31 further includes a shape recognition unit 32 and a dimension measuring unit 33. The shape recognition unit 32 forms a shape identification device. The dimension measuring unit 33 is used to measure the dimension of the delivery package 60 accommodated in the measurement receiving chamber 21 ( Fig. 3).

[0032] The shape recognition unit 32 compares external shape information acquired by the first camera 23 with information about a plurality of types of external shapes stored in the storage unit 35 to perform identification, and recognizes an external shape that matches or approximates the external shape information from the first camera 23. Therefore, the shape recognition unit 32 recognizes an external shape that matches or approximates the external shape information from the first camera 23, for example, by using a model for identifying external shapes of delivery packages generated by learning based on machine learning.At this time, as the external shape identification model, a model in which images of patterns are learned based on deep learning using a neural network with a multi-layer structure in which an input layer, two or more intermediate layers, and an output layer are connected by neurons can be used. In this case, by inputting the external shape information from the first camera 23 to the input layer, it is also possible to recognize an external shape represented by sample external shape information stored in the storage unit 35 and matching or approximating the external shape information from the first camera 23.For example, when a rectangular parallelepiped delivery package is imaged by the first camera 23, the outer shape of the delivery package can be recognized as a rectangular parallelepiped shape by using a matching or approximating outer shape from the outer shape information stored in the storage unit 35. For example, in the shape recognition unit 32, when a degree of approximation is determined by a predetermined approximation evaluation method, it is possible to recognize an outer shape from the outer shape information from the storage unit to which the outer shape information from the first camera has a predetermined degree of approximation or more.

[0033] Furthermore, the shape recognition unit 32 may perform filtering processing to remove noise in an image captured by the first camera 23 before comparing the image with the outer shape information stored in the storage unit 35. Thereafter, the image may be subjected to binarization processing to clarify the edge of the outer shape of the delivery package 60 ( Fig. 3).

[0034] Furthermore, the arithmetic processing unit 31 may be configured to include a shape learning unit 34. For example, the shape learning unit 34 may additionally store in the storage unit 35 external shape information from the first camera 23 corresponding to an external shape recognized by the shape recognition unit 32. As a result, an external shape that has not been stored in advance in the storage unit 35 can be easily identified and recognized through learning.

[0035] Furthermore, the shape learning unit 34 may perform learning by additionally inputting information of a pattern image associated with an external shape from the interface 36 and additionally storing a model for identifying external shapes of delivery packages 60 or 60a, which was created through the learning, in the storage unit 35. At this time, the external shape identification model may be a neural network with a multi-layer structure that performs learning through deep learning.

[0036] The dimension measuring unit 33 is provided as a part of the measuring device 20 in the arithmetic processing unit 31 and measures the dimension of the delivery package 60, for example, a dimension used for volume measurement, based on an external shape detected by the shape recognition unit 32. Note that the distance sensor 40 ( Fig. 4) can be used for dimension measurement. At this time, the dimension measuring unit 33 measures, for example, a dimension used for volume measurement of the delivery package 60 based on the external shape recognized by the shape recognition unit 32, the distance detected by the distance sensor 40, and the information of the image captured by the first camera 23.

[0037] Fig. 5 is a flowchart showing shape recognition and dimension measurement in the embodiment. In a state where the sender places a delivery package 60 in the measurement receiving chamber 21 and closes the door 22, the shape recognition processing of the delivery package 60 is started in step S1. In the shape recognition processing, the shape recognition unit 32 compares external shape information from the first camera 23 with external shape information stored in the storage unit 35 to perform identification, and recognizes an external shape that matches or approximates the external shape information from the first camera 23. For example, in a Fig. In the example shown in FIG. 5, the outer shape is recognized as one of a rectangular parallelepiped shape (box-like shape), a cylindrical shape, a predetermined shape, and other shapes. The rectangular parallelepiped shape refers to shapes including a cube. The "predetermined shape" includes specific shapes other than the rectangular parallelepiped shape and the cylindrical shape, which are stored in advance in the storage unit 35. The predetermined shape is, for example, an elongated shape with a predetermined length or more, and includes, for example, a shape of a body in which a golf bag with golf clubs accommodated therein is packed, a shape of a ski bag with skis accommodated therein, etc. "Other shapes" are shapes other than the rectangular parallelepiped shape, the cylindrical shape, and the predetermined shape.

[0038] Next, in step S2, dimension measurement processing is performed. In dimension measurement processing, dimension measurement and volume calculation are performed using different measurement methods according to the external shape recognized by the shape recognition unit 32. For example, if the external shape is recognized as a rectangular parallelepiped, the volume is calculated by width W × depth D × height H. For this reason, the dimension measurement unit 33 measures the width W, depth D, and height H of the delivery package and then calculates the volume based on the measured dimensions.

[0039] If the outer shape is recognized as the cylindrical shape, the volume is calculated by 2 × π × radius r × height H. Therefore, the dimension measuring unit 33 measures the radius r and the height H of the delivery package and then calculates the volume based on the measured dimensions.

[0040] When the external shape is recognized as the predetermined shape, a specific service is performed. For example, the specific service measures only the entire length as the dimension, or assumes and sets a dimension measurement value and a volume calculation value as a predetermined dimension and volume, respectively.

[0041] If the outer shape is recognized as a different shape, the volume measurement method is simply determined as width W × depth D × height H. Then, the dimension measuring unit 33 measures the width W, depth D, and height H of the delivery package and calculates the volume based on the measured dimensions. Alternatively, if the outer shape is recognized as a different shape, processing may be performed as error processing to guide the sender to a counter.

[0042] Next, in step S3, the results measured and calculated in the dimension measurement processing are displayed on the operation display unit 17. At this time, if processing corresponding to "specific service" or "other shapes" is executed in the dimension measurement processing, the fact that the processing was executed and a corresponding value are displayed. Note that if processing corresponding to "other shapes" is executed, a guidance message for guiding the sender to the counter of the delivery company ("Please come to the counter" or the like) is displayed as error processing on the operation display unit 17. The end of step S3 causes the shape recognition and dimension measurement processing to end. When the measurement result and the calculation result are displayed on the operation display unit 17, a delivery fee corresponding to the calculation result can be displayed.When the “specific service” is executed in the dimension measurement processing, a transmission fee corresponding to the service may be displayed on the operation display unit 17.

[0043] Fig. 6 is a flowchart showing the shape recognition and dimension measurement in the embodiment in more detail. Fig. 7 is a perspective view showing an example of the measurement receiving chamber 21 of the measuring device 20.

[0044] In step S11 of Fig. 6, the illumination unit 24 provided in the measurement receiving chamber 21 is turned on, so that the measurement receiving chamber 21 becomes bright. In step S12, the delivery package (measurement target) 60 is imaged by the first camera 23. At this time, the distance to the delivery package 60 can be detected by the distance sensor 40.

[0045] At this time, as in Fig. 7, first lines L1, second lines L2, and third lines L3 are drawn as a plurality of reference lines on a first side surface portion 80, a second side surface portion 81, and a bottom surface portion 82 constituting the measurement receiving chamber 21. The first side surface portion 80, the second side surface portion 81, and the bottom surface portion 82 intersect at a corner portion 84 of a reference point at the other end in the depth direction Y of the measurement receiving chamber 21. The first side surface portion 80 is located at the other end in the depth direction Y. The plurality of reference lines serve to improve the measurement accuracy of the dimensional measurement and are similar to a configuration described in Patent Literature 1.

[0046] Specifically, the first side surface portion 80 intersects the lower surface portion 82 at a first boundary line A1. The second side surface portion 81 intersects the lower surface portion 82 at a second boundary line A2. The first side surface portion 80 and the second side surface portion 81 intersect at a third boundary line A3. The plurality of first lines L1 drawn on the first side surface portion 80 are straight lines arranged at equal intervals parallel to the first boundary line A1. The plurality of second lines L2 drawn on the second side surface portion 81 are straight lines arranged at equal intervals parallel to the second boundary line A2.Each of the plurality of third lines L3 drawn on the lower surface portion 82 has an L-shape bent at a right angle at the center thereof, and first straight line portions L3a located on one end side and parallel to the first boundary line A1 and second straight line portions L3b located on the other end side and parallel to the second boundary line A2 are connected to each other. For the respective lines L1, L2, and L3, colors different from the respective colors of the first side surface portion 80, the second side surface portion 81, and the lower surface portion 82 are preferably used, and the luminance values ​​are far apart from each other. As shown in the figures described later. Fig. 8 and Fig. 9, the delivery packages 60 and 60a are preferably arranged on the lower surface section 82 so that they are pushed in the direction of the third boundary line A3 in the measurement receiving chamber 21.

[0047] In step S13 of Fig. 6, the shape recognition unit 32 captures an image taken by the first camera 23. Fig. 8 is a view showing an edge image created based on an image captured by the first camera 23 when the delivery package 60 is a rectangular parallelepiped. Fig. 9 is a view showing an edge image created based on an image captured by the first camera 23 when the delivery package 60a has a cylindrical shape.

[0048] For example, if the delivery package 60 is a rectangular parallelepiped, an image is obtained which is Fig. 8. For example, if the delivery package 60a has a cylindrical shape, an image is obtained which is Fig. 9 is shown.

[0049] An image of the plurality of first lines L1, second lines L2 and third lines L3 may be stored in advance in the storage unit 35 as mask data for performing mask processing, and the shape recognition unit 32 may perform the mask processing at a Fig. 8 or Fig. 9 using the mask data. As a result, the first lines L1, the second lines L2, and the third lines L3 can be deleted from the image to generate an image for shape recognition. Therefore, the edges of the outer shapes of the delivery packages 60 and 60a become clearer in the shape recognition images, so that the outer shapes are more easily identified. The edge image is generated by performing filter processing for removing noise on the image captured by the first camera 23. By performing binarization processing on the edge image, the edges of the outer shapes of the delivery packages 60 and 60a can be further clarified.

[0050] In step S14 of Fig. 6, the shape recognition unit 32 compares the outer shape information acquired by the first camera 23 with the plurality of types of outer shape information stored in the storage unit 35 to perform identification, and recognizes an outer shape that matches or approximates the outer shape information from the first camera 23.

[0051] In step S15 in Fig. 6, the control device 30 determines a dimension measuring method corresponding to the detected outer shape. For example, in the case of the Fig. 8, when its outer shape is recognized as a rectangular parallelepiped shape, the dimensional measurement method of the volume is determined as width W × depth D × height H.

[0052] Next, in step S16, the dimension measuring unit 33 measures a target dimension according to the determined dimension measuring method. For example, if the dimension measuring method of the volume is set as width W × depth D × height H for the Fig. 8, the dimension measuring unit 33 determines the width W, the depth D, and the height H of the delivery package 60. At this time, the dimension measuring unit 33 may measure the dimension using an image in which the first lines L1, the second lines L2, and the third lines L3 remain in the edge image captured by the first camera 23. For example, in the image shown in Fig. 8, edges and corner points of the delivery package 60 are detected, and the height H of the delivery package 60 can be measured depending on where a line E1 connecting the upper corner points P1 and P6 of the corner points P1 to P6 is located with respect to the plurality of first lines L1 on the first side surface portion 80. In Fig. 8, for example, since it is detected that the line E1 connecting the vertices P1 and P6 is located between second and third first lines L1 from the bottom, the height H is measured as a specific range. At this time, the height H of the delivery package 60 can be measured depending on where a line E2 connecting the upper vertices P5 and P6 from the vertices P1 to P6 is located with respect to the plurality of second lines L2 on the second side surface portion 81. Further, similar to the measurement of the height H, the width W of the delivery package 60 can be measured depending on where a line E3 connecting the vertices P3 and P4 is located with respect to the first straight line portions L3a of the plurality of third lines L3 on the lower surface portion 82.Furthermore, similar to the measurement of the height H, the depth D of the delivery package 60 can be measured depending on where a line E4 connecting the vertices P2 and P3 is located with respect to the second straight line segments L3b of the plurality of third lines L3 on the lower surface portion 82.

[0053] In a case where it is recognized that the Fig. 9 has a cylindrical shape, when the dimension measurement method of the volume is determined as 2 × π × radius r × height H, the dimension measuring unit 33 measures the radius r and the height H of the delivery package 60. At this time, the dimension measuring unit 33 can measure the dimension by using an image in which the first lines L1, the second lines L2, and the third lines L3 remain in the edge image captured by the first camera 23. For example, edges and vertices of the delivery package 60a are determined in the Fig. 9, and the height H of the delivery package 60 can be measured depending on where an arc passing through the upper vertices P8 to P10 is located with respect to the plurality of first lines L1 on the first side surface portion 80. Further, the height H of the delivery package 60 can be measured depending on where the arc is located with respect to the plurality of second lines L2 on the second side surface portion 81. Furthermore, it is possible to estimate a circle when the direction of the line of sight is changed so that an ellipse including the arc passing through the vertices P8 to P10 becomes the circle, and to measure the radius r of the estimated circle.

[0054] Next, the dimension measuring unit 33 calculates the volumes of the delivery packages 60 or 60a ( Fig. 8 or Fig. 9) according to the specific dimensional measurement method based on the measured dimensions.

[0055] In step S18, the control device 30 causes the display unit to display the dimension measured by the dimension measuring unit 33 and the volume calculated by the dimension measuring unit 33.

[0056] According to the measuring device 20 and the delivery box 12, the shapes of the delivery packages 60 or 60a accommodated in the measurement receiving chamber 21 can be automatically identified. As a result, the sender of the delivery package does not need to input the shape of the delivery package 60, 60a, thereby increasing the convenience for the sender.

[0057] As described above, the measuring device 20 may include the distance sensor 40 for detecting the distances to the delivery packages 60 or 60a. At this time, the outer shape to be stored in the storage unit 35 may be associated with the distance information. Then, the shape recognition unit 32 associates the outer shape information from the first camera 23 with the distance information from the distance sensor 40 and compares the outer shape information from the first camera 23 with the outer shape information stored in the storage unit 35 to perform identification. At this time, the shape recognition unit 32 recognizes an outer shape that matches or approximates the outer shape information from the first camera 23. As a result, the recognition accuracy of the outer shape is further improved.Furthermore, the accuracy of the dimensional measurement of the delivery packages 60 or 60a can be improved by using the detection result of the distance sensor 40. The above description was given for the case where the first lines L1, the second lines L2, and the third lines L3 are drawn in the measurement receiving chamber 21, but these lines can be omitted. In the case where the distance sensor 40 is used, the dimensions of the delivery packages 60 and 60a can be easily measured using the detection result of the distance sensor 40, even if the lines of the measurement receiving chamber 21 are omitted.Furthermore, even when the delivery packages 60 or 60a are moved from the third line L3 side, which is the other end in the depth direction Y, to the opening side and are arranged in any orientation in the measurement receiving chamber 21, it is easy to recognize the outer shapes of the delivery packages 60 and 60a using the captured image from the first camera 23 and the detection result of the distance sensor 40. At the same time, it is easy to measure the dimensions of the outer shape.

[0058] Further, as the measurement receiving chamber, a chamber in which the above-described lines L1 to L3 are not provided and rectangular frames are drawn on peripheral edge portions of the first side surface portion 80, the second side surface portion 81, and the bottom surface portion 8 may be used. The interior of the measurement receiving chamber is imaged in advance by the camera in a state where no object such as a delivery package is present, and based on the image data, it is possible to generate correction data for correcting distortion, etc., of the external shape of a measurement target in a captured image of the receiving chamber in which the measurement target is located.On the other hand, when the measurement receiving chamber in which first lines L1, second lines L2 and third lines L3 arranged at equal intervals as drawn in the above case is used, the accuracy of the dimensional measurement can be further improved by using the correction data.

[0059] Fig. 10 is a perspective view showing a measuring device 20a as a shape identification device according to another example of the embodiment. In the configuration of this example, the measuring device 20a includes a second camera 27, which is a second imaging unit. The measuring device 20a does not include a distance sensor. The configuration of the second camera 27 is similar to that of the first camera 23, and it is connected to the control device 30. The operation of the second camera 27 is controlled by the control device 30. The second camera 27 images the delivery package 60 in a direction different from a direction in which the first camera 23 performs imaging, for example, from the other end in the width direction X (a right end portion in Fig. 10), for example, at a corner portion of the upper portion on the opening side of the measurement receiving chamber 21. Information of an image captured by the second camera 27 is transmitted to the control device 30.

[0060] The shape recognition unit 32, with which the control device 30 is equipped, compares the outer shape information from the first camera 23 and the outer shape information from the second camera 27 with the outer shape information stored in the storage unit 35 to perform identification. Then, the shape recognition unit 32 recognizes an outer shape that matches or approximates the outer shape information from at least one of the first camera 23 and the second camera 27. For example, if the degree of approximation is determined by a preset approximation evaluation method, it is possible to recognize an outer shape from the storage unit 35 of outer shape information to which the outer shape information from one or both of the first camera 23 and the second camera 27 has a predetermined degree of approximation or more.

[0061] According to the above configuration, the recognition accuracy of the outer shape is improved. Furthermore, the accuracy of the dimension measurement of the delivery package 60 can be improved by using the outer shape information from the second camera 27. Furthermore, even when the delivery package 60 is picked up from the other end in the depth direction Y (the rear end of the paper surface of Fig. 10) is moved to the opening side and placed in any orientation in the measurement receiving chamber 21, it is easy to recognize the outer shape of the delivery package 60 using the captured images from the first camera 23 and the second camera 27. At the same time, it is easy to measure the dimensions of the outer shape. In this example, other configurations and operations are similar to the configurations of Fig. 1 to 9.

[0062] Note that in each of the examples described above, the case where the dimensions of the delivery package are measured and then its volume is determined was described. However, these examples are not limited to this method, and they can be configured to measure only the required dimensions according to the external shape of the delivery package, and not to determine its volume. For example, the delivery fee can be determined based on the measured required dimensions.

[0063] Note that in the configuration of each of the examples described above, the dimension measuring unit may be provided separately from the shape identification device. In this case, the dimension measuring unit measures the dimension of a measurement target based on an external shape recognized by the shape recognition unit 32 ( Fig. 4).

[0064] Furthermore, the delivery box 12 ( Fig.1) be configured such that the plurality of delivery receiving chambers include a plurality of types of receiving chambers with different sizes, and depending on the dimension of a delivery package measured by the measuring device 20, the door of a receiving chamber with the corresponding size is unlocked. At this time, the sender can place a delivery package in the unlocked receiving chamber and submit a delivery request. LIST OF REFERENCE SYMBOLS

[0065] 12 delivery box, 13 outer body, 14 delivery receiving chamber, 15 door, 17 operation display unit, 18 electric lock, 19 parcel sensor, 20, 20a measuring device, 21 measuring receiving chamber, 22 door, 23 first camera, 24 lighting unit, 26 cover plate unit, 27 second camera, 30 control device, 31 arithmetic processing unit, 32 shape recognition unit, 33 dimension measuring unit, 34 shape learning unit, 35 storage unit, 36 interface, 40 distance sensor, 60, 60a delivery parcel, 70 accounting unit, 71 communication network, 72, 73 terminal device, 74 management server, 80 first side surface portion, 81 second side surface portion, 83 lower surface portion, 84 corner portion.

Claims

[1] A delivery box (12) comprising a shape identification device (20) and an operation display unit (17), wherein the shape identification device (20) comprises: - a receiving chamber (14, 21) for receiving a measurement target object (60) so that the measurement target object (60) can be placed therein and taken out, wherein the measurement target object (60) is a delivery package (60); - a first imaging unit (23) for imaging the measurement target object (60) received in the receiving chamber (14, 21); and - an image processing unit (30) for performing image processing based on information of an image captured by the first imaging unit (23) to recognize an external shape of the measurement target object (60), the image processing unit (30) including: ◯ a storage unit (35) for storing in advance a plurality of types of pattern outer shapes corresponding to the measurement target object (60), ◯ a shape recognition unit (32) for comparing outer shape information from the first imaging unit (23) with outer shape information stored in the storage unit (35) to perform identification, and for recognizing an outer shape that matches or approximates the outer shape information from the first imaging unit (23) as one of a rectangular parallelepiped shape, a cylindrical shape, and an elongated shape with a predetermined length or more, which is different from the rectangular parallelepiped shape and the cylindrical shape, and o a dimension measuring unit (33) for measuring a dimension of the measurement target object (60) based on an external shape recognized by the shape recognition unit (32); wherein, when the shape recognition unit (32) recognizes the outer shape as the rectangular parallelepiped shape or the cylindrical shape, the dimension measuring unit (33) is configured to calculate a volume based on the measured dimension, and the delivery box (12) is configured to display a delivery fee corresponding to the calculation result on the operation display unit (17); and wherein, when the shape recognition unit (32) recognizes the outer shape as the elongated shape having the predetermined length or more, the delivery box (12) is configured to display a sending fee corresponding to a specific service on the operation display unit (17). [2] The delivery box (12) according to claim 1, wherein the image processing unit (30) is capable of additionally storing in the storage unit (35) the outer shape information from the first imaging unit (23) corresponding to the outer shape recognized by the shape recognition unit (32). [3] The delivery box (12) according to claim 1, wherein the image processing unit (30) is capable of additionally learning information of a pattern image associated with an external shape and additionally storing a model for identifying an external shape of the measurement target object (60) in the storage unit (35), the model being generated by learning. [4] The delivery box (12) according to claim 1, further comprising a distance sensor (40) for detecting a distance to the measurement target object (60), wherein an external shape to be stored in the storage unit (35) is associated with distance information, and wherein the shape recognition unit (32) associates external shape information from the first imaging unit (23) with distance information from the distance sensor (40), compares the external shape information from the first imaging unit (23) with the external shape information stored in the storage unit (35) to perform identification, and recognizes an external shape that matches or approximates the external shape information from the first imaging unit (23). [5] The delivery box (12) according to claim 1, further comprising a second imaging unit (27) for imaging the measurement target object (60) in a direction different from a direction in which the first imaging unit (23) performs imaging, and wherein the shape recognition unit (32) compares external shape information from the first imaging unit (23) and external shape information from the second imaging unit with the external shape information stored in the storage unit (35) to perform identification, and recognizes an external shape that matches or approximates the external shape information from at least one of the first imaging unit (23) and the second imaging unit.

Images