Weight determination system and weight determination program

Abstract

To provide a weight determination system which can determine the weight of a cargo using a simple configuration and is widely applicable.SOLUTION: A weight determination system is provided, comprising cargo information storage means configured to store cargo information including an image and the weight of a cargo for each cargo type, a smartphone 3 for capturing an image of a determination target cargo 2, and a management server 4 configured to determine the weight of the determination target cargo 2 on the basis of the captured image acquired by the smartphone 3 and the cargo information.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a weight determination system and a weight determination program for determining the weight of a load when loading the load onto a truck.

Background Art

[0002] For example, when loading a load such as a tower material onto a truck crane (unique vehicle), conventionally, it has been necessary for an operator to adjust the quantity so that the load weight is below a specified weight. Also, when packing the load prior to transporting it by helicopter, it has been necessary to check the weight of each load and the total weight and adjust so that the transport weight is below the specified weight.

[0003] Also, a suspended load re-calculation device is known that can improve the accuracy of the indicated load of a suspended load lifted by a crane (see, for example, Patent Document 1). This device has a detection unit that detects information for specifying the load of the suspended load of the crane, a calculation unit that calculates the suspended load weight based on the value detected by the detection unit, and an acquisition unit that acquires the boom length, and corrects the suspended load weight with a correction value corresponding to the boom length acquired by the acquisition unit.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Incidentally, when loading cargo with a truck crane or packing cargo for transport by helicopter, workers have traditionally had to check the weight of each item and adjust the total weight to stay below a specified weight. This is not only laborious and time-consuming, but also carries the risk of exceeding the weight limit due to checking errors. Furthermore, while the device described in Patent Document 1 can improve the accuracy of the indicated load for suspended loads, it requires a detection unit, calculation unit, acquisition unit, etc., making its configuration complex and not easily applicable to existing truck cranes. Moreover, it can only be applied to cranes used for lifting loads and cannot be used for packing that does not use a crane.

[0006] Therefore, the present invention aims to provide a weight determination system and weight determination program that can determine the weight of a package with a simple configuration and is widely applicable. [Means for solving the problem]

[0007] To solve the above problems, the invention of claim 1 comprises: a luggage information storage means for storing luggage information including an image and weight of each type of luggage; a photography means for photographing a luggage to be determined; and a weight determination means for determining the weight of the luggage to be determined based on the photographed image taken by the photography means and the luggage information. The aforementioned photographic means is a portable terminal equipped with a display unit, the determination result by the weight determination means is displayed on the display unit, the weight determination means sequentially determines the weight of multiple packages to be determined, outputs an alarm if the total weight exceeds a predetermined specified total weight, and the specified total weight can be input by the user. This is a weight determination system characterized by the following features.

[0008] According to this invention, when a photographing means photographs the luggage to be determined, the weight determination means determines the weight of the luggage based on the photographed image and the luggage information stored in the luggage information storage means.

[0009] The invention of claim 2 is a weight determination system according to claim 1, wherein the weight determination means outputs an alarm when the weight of the package to be determined exceeds a predetermined specified weight. death , The aforementioned specified weight can be entered by the user. It is characterized by the following:

[0010] The invention of claim 3 is characterized in that, in the weight determination system according to claim 1 or 2, the luggage information storage means stores the luggage information for each type and size of luggage.

[0011] The invention of claim 4 is characterized in that, in the weight determination system described in claims 1 to 3, the weight determination means uses a weight determination learning model that has been machine-trained based on past performance data, such that when the captured image and the luggage information are input, the weight of the luggage to be determined is output.

[0013] Claim 5 The invention provides a computer with a package information storage means that stores package information, including an image and weight of the package, for each type of package, and determines the weight of the package based on the captured image of the package to be determined and the package information. and output It functions as a means of determining weight. The weight determination means sequentially determines the weight of multiple packages to be determined, and outputs an alarm if the total weight exceeds a predetermined total weight, the predetermined total weight being inputtable by the user. This is a weight determination program characterized by the following features.

[0014] Claim 6 The invention is claimed 5 In the weight determination program described above, the weight determination means outputs an alarm when the weight of the package to be determined exceeds a predetermined specified weight. death , The aforementioned specified weight can be entered by the user. It is characterized by the following:

[0015] Claim 7 The invention is claimed 5 or 6 In the weight determination program described above, the luggage information storage means is characterized in that it stores the luggage information for each type and size of luggage.

[0016] Claim 8 The invention is claimed 5 to 7 The weight determination program described above is characterized in that the weight determination means uses a weight determination learning model that has been machine-trained based on past performance data, such that when the captured image and the luggage information are input, the weight of the luggage to be determined is output. [Effects of the Invention]

[0018] According to the invention described in claim 1 and claim 5 the weight of the luggage to be determined is determined based on the photographed image of the luggage to be determined, the image of the luggage stored in advance, and the weight. Therefore, it is possible to determine the weight of the luggage to be determined with a simple configuration. That is, it is possible to determine the weight of the luggage to be determined only by photographing the luggage to be determined, without measuring the weight of each piece of luggage to be determined or equipped with various devices. Moreover, since it is only necessary to photograph the luggage to be determined, it can be widely applied in various scenes.

[0019] Claim 3 and claim 7 According to the invention described in, since the image and weight are stored in advance for each type and size of the luggage, it is possible to more appropriately determine the weight of the luggage to be determined by considering not only the type but also the size of the luggage to be determined.

[0020] Claim 4 and claim 8 According to the invention described in, since the weight of the luggage to be determined is output using a machine-learned learning model for weight determination, it is possible to more appropriately determine the weight of the luggage to be determined.

[0021] Claim 2 and claim 6 According to the invention described in, when the weight of the luggage to be determined exceeds a predetermined specified weight, an alarm is output. Therefore, it is possible to prevent the lifting or conveyor transfer of luggage exceeding the predetermined specified weight in advance and ensure safety. That is, it is possible to surely prevent weight over due to a confirmation mistake.

[0022] Also, claim 1 and claim 5According to the invention described above, the weight of multiple loads to be judged is determined sequentially, and an alarm is issued if the total weight exceeds a predetermined specified total weight. This prevents the loading or transportation of loads exceeding the predetermined specified total weight onto trucks, thereby ensuring safety. In other words, it is possible to reliably prevent exceeding the total weight due to a mistake in checking. [Brief explanation of the drawing]

[0023] [Figure 1] This is a schematic diagram showing a weight determination system according to an embodiment of the present invention. [Figure 2] This figure shows the first guidance screen of the smartphone used for the weight determination system shown in Figure 1. [Figure 3] This figure shows the second (a) and third (b) guidance screens of the smartphone used in the weight determination system shown in Figure 1. [Figure 4] This figure shows the fourth guidance screen of the smartphone used for the weight determination system shown in Figure 1. [Figure 5] This figure shows the fifth guidance screen of the smartphone used in the weight determination system shown in Figure 1. [Figure 6] Figure 1 is a schematic block diagram showing the management server of the weight determination system. [Figure 7] Figure 6 is a data configuration diagram of the package information database on the management server. [Figure 8] Figure 7 shows the first example image (a) and the second example image (b) stored in the cargo information database. [Figure 9] Figure 6 is a functional block diagram showing the schematic configuration of the learning model for weight determination on the management server. [Modes for carrying out the invention]

[0024] The present invention will be described below based on the illustrated embodiments.

[0025] Figure 1 is a schematic diagram showing a weight determination system 1 according to an embodiment of the present invention. This weight determination system 1 is a system for determining the weight of a load 2 to be determined, and can be used in any scene or situation for determining the weight of the load 2. However, in this embodiment, the system will be mainly described below using the example of determining the weight of load 2 when loading multiple loads 2 to be determined onto a truck T. In contrast, it can also be applied when packing multiple loads 2 to be determined for transport by helicopter, when transporting one load 2 to be determined by a forklift, when lifting load 2 to be determined by a hoist, or when transporting load 2 to be determined by a conveyor belt.

[0026] Here, the designation "2" is used to indicate the load 2 to be judged, while the designation is omitted when referring to a load used as a reference when determining the weight of load 2 to be judged (reference load). Furthermore, although this embodiment mainly describes the case where the load 2 to be judged is a component of a steel tower, it is of course possible to use other loads as well.

[0027] The weight determination system 1 comprises a smartphone 3 that primarily functions as a means of taking pictures, and a management server 4 that primarily functions as a means of storing luggage information and a means of determining weight. It is possible to connect multiple smartphones 3 to one management server 4. In other words, multiple smartphones 3 can send weight determination requests, as described later, to the management server 4, and the management server 4 can return determination results to the multiple smartphones 3.

[0028] Smartphone 3 has a configuration equivalent to that of off-the-shelf, existing multi-functional mobile terminals, equipped with a camera function (shooting means) and a touch panel (display and input unit). It is carried by a worker M or similar person, who takes a picture of the package 2 to be judged and transmits it to the management server 4. Thus, this embodiment describes the case where worker M takes a picture using smartphone 3, but a device equipped with a camera function and communication function may also be installed to automatically photograph the package 2 to be judged and transmit it to the management server 4. Furthermore, the captured image of the package 2 to be judged may be a still image or a video, but this embodiment mainly describes the case of a still image.

[0029] Furthermore, this smartphone 3 has a weight determination app installed, which is an application program that guides the user on the screen (touch panel) through taking pictures of the luggage 2 to be judged and sending the captured images. This weight determination app is launched when a predetermined operation is performed, and first displays a first guidance screen (login screen) on the touch panel, as shown in Figure 2. This first screen has an input field for entering the specified total weight, and when the specified total weight, which is the maximum weight that can be loaded onto truck T, is entered into this input field and the "Start Weight Determination" button on the screen is pressed, a message is sent to the management server 4 indicating that the weight determination has started and the camera is activated.

[0030] As described above, in this embodiment, the specified total weight is entered and set on the smartphone 3, but it may also be set and registered in advance on the management server 4 depending on the weight determination scene and the truck T used. In addition, depending on the weight determination scene, it may also be possible to input a specified weight (specified individual weight) in addition to the specified total weight. This specified weight is the upper limit weight of one load 2 to be determined, and is, for example, the upper limit of the load that can be lifted by the crane (Unic) T1 of the truck T.

[0031] Next, as shown in the second guidance screen in Figure 3(a), when the luggage 2 to be judged is photographed by the camera, the third guidance screen in Figure 3(b) is displayed on the touch panel. Here, the luggage 2 may be photographed in any state; for example, as shown in Figure 3(a), it may be suspended by the crane T1, or it may be in a state before being suspended by the crane T1.

[0032] The third guidance screen displays the captured image, a "Weight Determination" button, a "Retake Shoot" button, and other buttons. When the "Retake Shoot" button is pressed, the user returns to the second guidance screen shown in Figure 3(a), allowing for a retake of the target package 2 (re-shooting). On the other hand, when the "Weight Determination" button is pressed, a weight determination request is sent to the management server 4, including the entered total weight (and the specified weight if one was entered), the captured image of the target package 2, and the identification information (IP address) of the smartphone 3.

[0033] Next, as described later, the weight of the target package 2 is determined by the management server 4, and when the management server 4 receives a determination result that is less than or equal to the total weight, the fourth guidance screen in Figure 4 is displayed on the touch panel. This determination result includes the weight of the target package 2 for which the weight determination request was sent (the weight of the package immediately before), the cumulative weight (total weight) of all target package 2 determined since the "Start Weight Determination" button was pressed, and the remaining weight that can be loaded by subtracting this total weight from the specified total weight. The fourth guidance screen displays these weights: the weight of the package immediately before, the total weight, and the remaining weight, as well as buttons such as "Continue Weight Determination" and "End".

[0034] Then, when the "Continue weight judgment" button is pressed, the camera is activated and the system returns to the second guidance screen in Figure 3(a), allowing the system to photograph the next item to be judged (item 2), and the process described above is repeated. When the "End" button is pressed, a message indicating the end of weight judgment is sent to the management server 4 and the weight judgment application is terminated.

[0035] On the other hand, when the system receives the result of the total weight overload determination from the management server 4, the fifth guidance screen shown in Figure 5 is displayed on the touch panel. This determination result includes the weight of the previous shipment, the total weight, and the specified total weight. The fifth guidance screen displays these values, along with a warning message "Warning: Specified total weight exceeded!" and an "End" button. In other words, the system sequentially determines the weight of multiple shipments 2 to be judged, and outputs a warning if their total weight exceeds the specified total weight. Furthermore, an alarm sound is emitted simultaneously with this display. This allows workers M to understand that the previous shipment 2 cannot be loaded onto truck T due to the total weight overload. When the "End" button is pressed, a message indicating the end of the weight determination is sent to the management server 4, and the weight determination application is terminated.

[0036] Here, if the specified weight is entered as described above, as will be explained later, each time the weight of one package 2 to be judged is judged, it is determined whether or not the weight is less than or equal to the specified weight. If it is greater than the specified weight, the management server 4 receives a weight overload judgment result. In this case as well, a similar alarm is output. Thus, an alarm is also output if the weight of the package 2 to be judged exceeds the predetermined specified weight. At this time, a "Continue weight judgment" button and an "End" button, similar to the fourth guidance screen, are displayed on the touch panel, making it possible to judge the weight of the next package 2 to be judged or to end the weight judgment application.

[0037] The management server 4 is located in the management center C and is a server computer used to determine the weight of the luggage 2 to be judged. As shown in Figure 4, it mainly comprises a storage unit 41, a communication unit 42, a weight determination task (weight determination means) 43, a learning task 44, and a central processing unit 45 that controls these. In this embodiment, the case in which the weight of the luggage 2 to be judged is determined by the management server 4 is described, but it may also be determined by a smartphone 3, i.e., a camera, in which case the management server 4 can be omitted.

[0038] The memory unit 41 mainly comprises a luggage information database (luggage information storage means) 411, a weight determination learning model 412, and a weight determination performance database 413. Here, the luggage information database 411 will be described, and the weight determination learning model 412 and the weight determination performance database 413 will be described later.

[0039] The luggage information database 411 is a database that stores luggage information, including the image and weight of each reference luggage, for each type and size of reference luggage. In other words, it is a database that stores information about reference luggage to be used as a reference when determining the weight of luggage 2 to be judged. As shown in Figure 7, for each luggage ID 4111, the type 4112, size 4113, image 4114, weight 4115, and other information 4116 are stored.

[0040] The package ID 4111 stores identification information for the reference package and its contents, and the type 4112 stores the type of the reference package. For example, the main column members, diagonal members, and horizontal members of a steel tower may be stored, along with the type of material, such as angle steel, hollow steel pipe, or concrete-filled steel pipe. The size 4113 stores the size and dimensions of the reference package, and the image 4114 stores an image of the reference package. For example, as shown in Figure 8(a), an image of the main column member of a steel tower made of angle steel is stored, and as shown in Figure 8(b), an image of the main column member of a steel tower made of hollow steel pipe is stored.

[0041] Here, the image is one that facilitates the determination of the weight of the cargo 2 to be determined in the weight determination task 43 described later. For example, an image of cargo 2 in a similar state to the photographed image of cargo 2 (the assumed state of cargo 2 at the time of photography, for example, when it is suspended by crane T1 or before it is suspended) is stored, and there may be one or more such images. The weight 4115 stores the weight of the reference cargo. That is, the type, size, and weight of the reference cargo in the image are stored as described above. In this way, the cargo information database 411 stores information about cargo (reference cargo) related to cargo 2 to be determined.

[0042] The communication unit 42 is an interface for communicating with the outside world, and for example, it receives weight determination requests from each smartphone 3 and transmits determination results to each smartphone 3.

[0043] The weight determination task 43 is a task program that determines and estimates the weight of the target package 2 based on the image captured by the smartphone 3 and the package information in the package information database 411. Specifically, it is activated when a weight determination request is received from the smartphone 3, and it performs image analysis on the image of the target package 2 included in the weight determination request and compares it with the image 4114 of each package ID 4111 stored in the package information database 411 to determine the weight of the target package 2 based on the weight 4115.

[0044] Specifically, the system identifies image 4114, which matches or closely resembles the image of the item to be judged (item 2), and determines the weight of item 2 from the weight 4115 of this image 4114. In this process, the size of item 2 is determined from the image of item 2. If the size is smaller than the size 4113, the difference is subtracted from the weight 4115. If the size of item 2 is larger than the size 4113, the difference is added to the weight 4115 to determine the weight of item 2.

[0045] This weight determination task 43 uses a machine learning model 412 for weight determination, which is based on past performance data, to output the weight of the target package 2 when a photograph of the target package 2 and package information of a reference package are input. This weight determination model 412 is created by the learning task 44.

[0046] In other words, the learning task 44 creates a weight determination learning model 412 using a known machine learning algorithm such as a neural network, based on past performance data recorded and stored in the weight determination performance database 413. This weight determination performance database 413 is a database in which performance data is recorded and stored, including the weight of the target luggage 2 determined by experts and experienced personnel in image analysis and weight determination, based on the captured image of the luggage 2 to be determined as input information and luggage information from the luggage information database 411, as well as the results of actually measuring the weight of the target luggage 2. The past performance data includes data created based on the actual captured image of the target luggage 2, luggage information from the luggage information database 411, and weights actually determined or measured by experts in image analysis, as well as data created through pre-training, etc.

[0047] As shown in Figure 9, this learning task 44 uses machine learning and deep learning with a neural network to create a neural network based on actual data recorded in the weight determination database 413. For example, the input layer is the image of the luggage 2 to be determined and the luggage information from the luggage information database 411, the output layer is the weight of the luggage 2 to be determined, and the hidden layer is the analysis process from the input layer to the output layer. Then, the learning task 44 uses the actual data of the weight determination learning model 412 as training data to learn various parameters in the hidden layer. In other words, the learning task 44 learns various parameters in the hidden layer so that the weight of the luggage 2 to be determined can be determined appropriately based on the image of the luggage 2 to be determined and the luggage information from the luggage information database 411.

[0048] Once the weight of the target package 2 is determined in this way, the cumulative weight (total weight) of all target package 2 that have been determined since the "Start Weight Determination" button was pressed on smartphone 3 is calculated, and it is determined whether this total weight exceeds the specified total weight received from smartphone 3. If it does not exceed the specified total weight, the determination result that the total weight is less than or equal to the specified total weight is sent to smartphone 3, which includes the weight of the most recently determined package 2 (the weight of the most recently determined package), the above total weight, and the remaining weight that can be loaded by subtracting the total weight from the specified total weight. On the other hand, if it exceeds the specified total weight, the determination result that the total weight is over, which includes the weight of the most recently determined package, the total weight, and the specified total weight, is sent to smartphone 3 (an alert is output).

[0049] Furthermore, when the specified weight is entered into smartphone 3, the system determines whether the weight of each item 2 being judged is below the specified weight. If the weight exceeds the specified weight, the system sends the weight overload result, including the determined weight, to smartphone 3 (outputs an alert).

[0050] To determine the weight of the cargo 2 to be weighed using the weight determination system 1 configured in this way, one simply needs to take a picture of the cargo 2 with a smartphone 3 and send the image to the management server 4. For example, when loading multiple cargo 2 to be weighed onto a truck T, one can launch the weight determination app on the smartphone 3 and, following the instructions displayed on the touch panel, sequentially take pictures of the cargo 2 and send the images to the management server 4. As a result, the management server 4 sequentially determines the weight of the multiple cargo 2, and the determination results, including the weight of each cargo 2, the total weight, and the remaining weight, are sent to and displayed on the smartphone 3.

[0051] As described above, this weight determination system 1 determines the weight of the luggage 2 based on a photographed image of the luggage 2 to be determined and a pre-stored image and weight of a reference luggage. Therefore, it is possible to determine the weight of the luggage 2 with a simple configuration. In other words, it is possible to determine the weight of the luggage 2 simply by photographing it, without having to weigh each luggage 2 individually or equipping it with various devices. Moreover, because it only requires photographing the luggage 2, it can be widely applied in a variety of scenes and situations.

[0052] Furthermore, since images and weights are pre-stored for each type and size of reference luggage, it becomes possible to determine the weight of luggage 2 more accurately by considering not only its type but also its size.

[0053] Furthermore, since the weight of the target package 2 is output using the machine learning-developed weight determination model 412, it becomes possible to determine the weight of the target package 2 more accurately.

[0054] Furthermore, if the weight of the item being checked (item 2) exceeds a predetermined weight limit, an alarm is issued. This prevents lifting or conveying of item 2 exceeding the predetermined weight limit, thereby ensuring safety. In other words, it is possible to reliably prevent weight overruns due to verification errors.

[0055] Furthermore, by sequentially checking the weight of multiple target items (item 2), an alarm is issued if the total weight exceeds a predetermined limit. This prevents the loading or transportation of items exceeding the predetermined limit onto truck T, thereby ensuring safety. In other words, it is possible to reliably prevent exceeding the total weight limit due to verification errors.

[0056] Furthermore, the smartphone 3 is equipped with a weight determination app that guides users through the process of taking photos of the item 2 to be determined and sending the captured images via a touch panel. This makes it easy and accurate for anyone to take photos of the item 2 to be determined and send the captured images by following the instructions on the screen.

[0057] Although embodiments of this invention have been described in detail above, the specific configuration is not limited to these embodiments, and any design changes, etc., that do not depart from the gist of this invention are also included. For example, in the above embodiments, the case in which the object to be judged 2 is a component of a steel tower was mainly described, but it is of course possible to use other objects as well. Furthermore, multiple objects to be judged 2 may be photographed simultaneously, in which case the weight of each object 2 included in the photographed image should be determined individually.

[0058] Alternatively, the management server 4 and weight determination system 1 described above may be configured by installing the following weight determination program on a general-purpose computer. That is, the weight determination program causes the computer to function as a luggage information storage means (luggage information database 411) that stores luggage information including an image and weight of each luggage for each type and size of luggage, and a weight determination means (weight determination task 43) that determines the weight of the luggage 2 to be determined based on the captured image and luggage information of the luggage 2 to be determined. The weight determination means uses a weight determination learning model 412 that has been machine-learned based on past performance data so that when the captured image and luggage information are input, the weight of the luggage 2 to be determined is output. The weight determination means outputs an alarm if the weight of the luggage 2 to be determined exceeds a predetermined specified weight, and further determines the weight of multiple luggage 2 to be determined sequentially and outputs an alarm if their total weight exceeds a predetermined specified total weight. [Explanation of Symbols]

[0059] 1. Weight determination system 2. Items subject to evaluation 3. Smartphone (Method of taking photos) 4. Management Server 411. Luggage Information Database (Luggage Information Storage Method) 412 Learning models for weight determination 413 Weight Determination Performance Database 43. Weight determination task (weight determination means) M worker T-Track

Claims

1. A package information storage means that stores package information, including an image and weight of each package, for each type of package, A means of photographing the luggage to be judged, A weight determination means that determines the weight of the luggage to be determined based on the captured image taken by the aforementioned photographic means and the luggage information, Equipped with, The aforementioned photographic means is a portable terminal equipped with a display unit, and the determination result from the weight determination means is displayed on the display unit. The weight determination means sequentially determines the weight of multiple packages to be determined, and outputs an alarm if the total weight exceeds a predetermined total weight. The aforementioned total weight is user-inputtable. A weight determination system characterized by the following features.

2. The weight determination means outputs an alarm if the weight of the package to be determined exceeds a predetermined specified weight. The aforementioned specified weight can be entered by the user. The weight determination system according to feature 1.

3. The aforementioned luggage information storage means stores the luggage information for each type and size of luggage. A weight determination system according to either claim 1 or 2.

4. The weight determination means uses a machine learning model for weight determination, which is based on past performance data, so that when the captured image and the luggage information are input, the weight of the luggage to be determined is output. A weight determination system according to any one of claims 1 to 3.

5. Computers, A package information storage means that stores package information, including an image and weight of each package, for each type of package, A weight determination means that determines and outputs the weight of the package to be determined based on the captured image of the package and the package information. To make it function as, The weight determination means sequentially determines the weight of multiple packages to be determined, and outputs an alarm if the total weight exceeds a predetermined total weight. The aforementioned total weight is user-inputtable. A weight determination program characterized by the following features.

6. The weight determination means outputs an alarm if the weight of the package to be determined exceeds a predetermined specified weight. The aforementioned specified weight can be entered by the user. The weight determination program according to feature 5.

7. The aforementioned luggage information storage means stores the luggage information for each type and size of luggage. A weight determination program according to any one of claims 5 or 6.

8. The weight determination means uses a machine learning model for weight determination, which is based on past performance data, so that when the captured image and the luggage information are input, the weight of the luggage to be determined is output. A weight determination program according to any one of claims 5 to 7.

Images