Discrimination device

Abstract

Even if the models are shared, the work based on setting and the like does not become complicated. A server device (20) acquires, by means of a recorder device (10), a mounting image captured by an imaging unit (11) mounted in a vehicle compartment (104) of a construction machine, and extracts, by means of a processing unit (22), a feature point image in the vehicle compartment (104) from the mounting image. Then, a processing unit (22) determines the type of the construction machine on the basis of a previously prepared image pool (23a) and the feature point image, and specifies a behavior estimation library for the recorder device (10) corresponding to the determined type.

Description

Technical Field

[0001] The present invention relates to a discriminating device for discriminating the type of construction machinery. Background Art

[0002] Driving recorders are becoming more common in vehicles such as automobiles, and are increasingly being used for purposes such as determining the cause of traffic accidents and managing driving of employees.

[0003] In construction machinery (also called working machinery) such as backhoes (also called hydraulic excavators), bulldozers, cranes, and dump trucks, it has also been proposed to install a driving recorder for the purpose of verifying accidents, etc. For example, Patent Document 1 describes the following: According to the movement of the lower traveling body, the upper swing body, or the working machine, a camera is selected by a camera selection device from a plurality of cameras mounted on the upper swing body in a manner capable of shooting in different directions, and a dynamic image is obtained by shooting during the movement of the lower traveling body, the upper swing body, or the working machine using the selected camera, and the dynamic image is stored in a dynamic image storage device.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2018-53537 Summary of the invention

[0007] Problems to be solved by the invention

[0008] For example, it has been proposed to estimate the motion (forward, backward, rotation, excavation, etc.) of a construction machine based on information obtained from a drive recorder or the like, and use the information to manage the operating conditions of the construction machine.

[0009] At present, most of the equipment such as driving recorders installed in construction machinery are designed according to the categories of construction machinery such as backhoes and bulldozers, and the parameters used to perform the above-mentioned actions are also set according to the categories. Therefore, the equipment such as driving recorders installed in construction machinery are produced in small quantities with multiple varieties, which is costly and complicated to distribute. On the other hand, when the models are made common, it is necessary to set the category and parameters when installing the equipment, which makes the work of installing the equipment complicated.

[0010] As an example of a problem to be solved by the present invention, even when the machine models are made common, operations such as setting will not become complicated.

[0011] Means for solving problems

[0012] In order to solve the above-mentioned problems, the invention recorded in technical solution 1 is characterized in that it comprises: an image acquisition unit, which acquires a captured image captured by a shooting unit installed in the cabin of the construction machinery; a feature point extraction unit, which extracts feature points inside and outside the cabin from the captured image; a category determination unit, which determines the category of the construction machinery based on pre-prepared information related to the feature points of each category of the construction machinery and the extracted feature points; and a determination unit, which determines the action parameters used by the shooting unit corresponding to the determined category.

[0013] The invention recorded in Technical Solution 6 is a discrimination method performed by a discrimination device for discriminating the category of construction machinery, and is characterized in that it includes the following steps: an image acquisition step, acquiring a captured image captured by a photographing unit installed in the cabin of the construction machinery; a feature point extraction step, extracting feature points inside and outside the cabin from the captured image; a category discrimination step, discriminating the category of the construction machinery based on pre-prepared information related to the feature points of each category of the construction machinery and the extracted feature points; and a determination step, determining the action parameters used by the photographing unit corresponding to the discriminated category.

[0014] The invention described in claim 7 is characterized in that the discrimination method described in claim 6 is executed on a computer as a discrimination program.

[0015] The invention described in claim 8 is characterized in that the determination program described in claim 7 is stored in a computer-readable storage medium. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic structural diagram of a system including a discrimination device and a determination device according to an embodiment of the present invention.

[0017] Figure 2 yes Figure 1 The functional structure diagram of the recorder device is shown.

[0018] Figure 3 yes Figure 1 The functional structure diagram of the server device shown.

[0019] Figure 4 yes Figure 1 A flowchart of the pre-processing actions of the system is shown.

[0020] Figure 5 yes Figure 1 A flow chart of the operation when the system shown is used.

[0021] Figure 6 is an example of an installation image.

[0022] Figure 7 is from Figure 6 The installation image shown is an example of an image in which the window frame portion is extracted. DETAILED DESCRIPTION

[0023] The following is a description of a discrimination device according to an embodiment of the present invention. The discrimination device according to an embodiment of the present invention obtains a captured image captured by a capturing unit installed in the cabin of a construction machine through an image acquisition unit, and extracts feature points inside and outside the cabin from the captured image through a feature point extraction unit. Then, the category discrimination unit discriminates the category of the construction machine based on pre-prepared information related to feature points of each category of the construction machine and the extracted feature points, and determines the action parameters for the capturing unit corresponding to the category discriminated by the determination unit. Thus, the category of the construction machine can be automatically discriminated based on the image captured by the capturing unit, and appropriate action parameters can be determined. Therefore, even when the machine models are made common, operations based on settings and the like will not become complicated.

[0024] In addition, the information related to the characteristic points may be characteristic points of seats or window frames provided in the vehicle cabin. The seats and window frames in the vehicle cabin are parts that differ greatly depending on the type of construction machinery, and when distinguishing the type, it is possible to distinguish more accurately.

[0025] In addition, the information related to the characteristic points may be further subdivided for each type of construction machinery in the category, thereby enabling identification of each manufacturer, each type, etc., and using more precise motion parameters.

[0026] In addition, the motion parameters are used to determine the motion of the construction machine. That is, as described above, the motion parameters can be used to determine the motion such as advancing and digging, and the operating status of the construction machine can be understood.

[0027] In addition, the imaging device may include the identification device, the imaging unit, and the recording unit for recording the image acquired by the image acquisition unit. Thus, in a device such as a driving recorder installed in a construction machine, the type of the construction machine can be automatically identified and appropriate motion parameters can be determined.

[0028] In addition, the information processing device may be configured to include a feature point acquisition unit that acquires the feature points extracted by the above-mentioned determination device and an accumulation unit that accumulates the acquired feature points. Thus, for example, by accumulating feature points in a server device, etc., it is possible to easily cope with new models, etc.

[0029] In addition, regarding the discrimination method of one embodiment of the present invention, the image acquisition step acquires a captured image captured by a camera installed in the cabin of the construction machinery, and the feature point extraction step extracts feature points inside and outside the cabin from the captured image. Then, the category discrimination step discriminates the category of the construction machinery based on the information related to the feature points of each category of the construction machinery prepared in advance and the extracted feature points, and the determination step determines the action parameters for the camera corresponding to the discriminated category. Thus, the category of the construction machinery can be automatically discriminated based on the image captured by the camera, and appropriate action parameters can be determined. Therefore, even when the machine model is made common, the operation based on the setting and the like will not become complicated.

[0030] In addition, the above-mentioned identification method may be executed by a computer as an identification program. Thus, the type of the construction machine can be automatically identified based on the image captured by the imaging unit using the computer, and appropriate operation parameters can be determined.

[0031] Furthermore, the above-mentioned discrimination program may be stored in a computer-readable storage medium, so that the program can be incorporated into a device or distributed as a single unit, and version upgrades can be easily performed.

[0032] Example

[0033] Reference Figure 1 and Figure 2 A discrimination device and a determination device according to an embodiment of the present invention will be described. Figure 1 It is a schematic structural diagram of a system including a discrimination device and a determination device of this embodiment.

[0034] Figure 1 The system 30 shown includes a recorder device 10 and a server device 20. The recorder device 10 is installed in a cabin 104 (also called a control room, a seat, or a cab, etc.) of a construction machine 100. The recorder device 10 and the server device 20 can communicate via a network N based on the Internet, a public line network, etc.

[0035] In this embodiment, the construction machine 100 is described as a backhoe, but may be other construction machines such as a bulldozer, a crane, and a dump truck. Figure 1 The engineering machine 100 shown has an upper slewing body 103 mounted on a crawler-type lower traveling body 101 so as to be rotatable around a slewing axis via a slewing mechanism 102. In addition, the upper slewing body 103 has a cabin 104 at its front left side, and has an excavation attachment consisting of a boom 105, an arm 106, and a bucket 107 at its front center. The cabin 104 is provided with a seat for an operator (also referred to as a driver or operator), and windows are provided on the front surface, the side surface, and the upper surface so that the operator can visually check the outside.

[0036] The recorder device 10 includes an imaging unit (camera) as described later, and images a range including substantially the front (a direction facing forward when the operator sits on a seat) from inside the vehicle cabin 104 . Figure 2 1 shows a functional configuration diagram of the recorder device 10. The recorder device 10 includes an imaging unit 11, a sensor 12, a control unit 13, a communication unit 14, and a recording unit 15.

[0037] The imaging unit 11 is constituted by, for example, a camera and has an imaging element, optical components, etc. As described above, the imaging unit 11 captures an image of a range including substantially the front from within the vehicle cabin 104. Image data captured by the imaging unit 11 is output to the control unit 13.

[0038] The sensor 12 includes, for example, a triaxial acceleration sensor, a GPS (Global Positioning System) receiver, etc. The triaxial acceleration sensor can detect vibration applied to the recorder device 10. In addition, the GPS receiver can acquire information on the current position and time.

[0039] The control unit 13 is composed of, for example, a microprocessor having a CPU (Central Processing Unit), a memory, etc. The control unit 13 is responsible for overall control of the recorder device 10. In addition, the control unit 13 transmits the image data captured by the imaging unit 11 when the recorder device 10 is installed or started to the server device 20 via the communication unit 14.

[0040] The communication unit 14 includes, for example, a wireless communication circuit and an antenna, and performs wireless communication with the network N. The communication unit 14 communicates with the server device by performing wireless communication with the network N, and transmits image data captured by the image capturing unit 11 .

[0041] The recording unit 15 is composed of, for example, a memory card slot and a memory card. The recording unit 15 records the image data captured by the imaging unit 11 in the memory card in sequence. That is, the image of the construction machine 100 in motion can be recorded in the recording unit 15. In addition to the image data, the information detected by the sensor 12 can also be recorded in the memory card in association with the image data. In addition, the recording unit 15 is not limited to a memory card, and can also be a non-volatile internal memory or the like.

[0042] The server device 20 is constituted by a well-known server computer. Figure 3 2 is a functional block diagram of the server device 20. The server device 20 includes a communication unit 21, a processing unit 22, and a storage unit 23.

[0043] The communication unit 21 is constituted by, for example, a network interface of a computer, etc. The communication unit 21 receives image data and the like transmitted from the recorder device 10 .

[0044] The processing unit 22 is composed of, for example, a microprocessor including a CPU of a computer, a memory, etc. As described later, the processing unit 22 determines the type of the construction machine on which the recorder device 10 is mounted based on the image data acquired by the communication unit 21. In addition, as described later, the processing unit 22 determines whether the mounting state of the recorder device 10 is appropriate based on the image data acquired by the communication unit 21.

[0045] The storage unit 23 is composed of a nonvolatile storage medium such as a hard disk, for example. The storage unit 23 stores various programs and data executed by the processing unit 22. In the present embodiment, the storage unit 23 stores an image pool 23a.

[0046] The image pool 23a is obtained by accumulating (saving) images representing information related to the feature points of the construction machinery within the range captured by the imaging unit 11 for each category and model. That is, the image pool 23a is information related to the above-mentioned feature points of each category and model of the construction machinery prepared in advance. In the present embodiment, the information related to the feature points is an image (feature point image) from which the window frame portion of the vehicle cabin 104 is extracted. The feature point image can be the entire window frame or only the columnar column portion. Alternatively, it can be only a roughly horizontal portion, but it is preferred that the portion that can identify each model be used.

[0047] Next, refer to Figure 4 and Figure 5 The flowchart of the above structure system 30 (discrimination method and determination method) is explained. Figure 4 and Figure 5 The flowchart is configured as a program executed by a computer having a CPU or the like as a discrimination program and a determination program.

[0048] Figure 4 1 is a flowchart of preprocessing. Preprocessing is a process of generating images accumulated (stored) in the above-mentioned image pool 23a. First, an installation image is captured in the recorder device 10 (step S101). The installation image refers to an image captured by the imaging unit 11 after the recorder device 10 is installed on the construction machine 100. The captured installation image is sent to the server device 20 via the network N.

[0049] Next, the processing unit 22 of the server device 20 generates a feature point image from the installation image (step S102). The feature point image corresponds to information related to the feature points of the construction machine. Figure 6 and Figure 7 The feature point image will be described.

[0050] Figure 6This is an image captured by the recorder device 10 installed on the upper left rear of the seat in the cabin 104 of a backhoe excavator manufactured by a certain construction machinery manufacturer, and is an example of the above-mentioned installation image. When acquiring the installation image, the processing unit 22 extracts the window frame portion from the installation image. Figure 7 is from Figure 6 An example of an image where the window frame is extracted. Figure 7 The image is simplified for illustration purposes. Figure 7 In the embodiment, the window frame is composed of members P1 to P3 extending in a substantially vertical direction and members F1 to F5 extending in a substantially horizontal direction.

[0051] The processing unit 22 uses a well-known method such as a machine learning model and template matching to obtain Figure 6 The image shown is generated Figure 7 The image shown. That is, in this embodiment, the window frame portion of the vehicle cabin 104 becomes a characteristic point. In addition, it is not limited to the entire window frame in the shooting range, and it may be only the parts P1 to P3 extending in a substantially vertical direction, or only the parts F1 to F5 extending in a substantially horizontal direction. Moreover, it may be only the parts P1 and P2 in the front, the part F5 on the upper side of the window on the side when viewed from the operator, etc., which are parts with characteristic shapes depending on the type and model of the construction machinery.

[0052] In addition, in the present embodiment, the feature point is set as the window frame or a part thereof, but it may also be a seat (seat) provided in the vehicle cabin 104. The seat may also be a part. Alternatively, it may be a part of the exterior of the vehicle cabin such as an excavation attachment seen from the window that has a characteristic shape based on the construction machinery. That is, the feature point only needs to be inside or outside the vehicle cabin 104. In a general driving recorder installed in a car or the like, the window frame or the like does not enter the shooting range, but in the case of construction machinery, the window frame or the like is mostly included in the shooting range, so it is effective to use it as a feature point.

[0053] Next, the processing unit 22 accumulates the feature point image generated in step S102 in the image pool 23a (step S103). When accumulating, the generated feature point image is added with information on the type (backhoe, bulldozer, etc.) of the construction machinery, the model (manufacturer name, model, etc.), and the installation angle of the recorder device 10. The information on the type, model, and installation angle can be obtained from the recorder device 10 or manually added.

[0054] In the present embodiment, the server device 20 performs the preprocessing, but the preprocessing may be performed by another device or the like, and the execution result may be stored in the storage unit 23 as the image pool 23 a.

[0055] Next, refer to Figure 5The operation during use is described in the flowchart of FIG. The use time here refers to the normal use time after the recorder device 10 is installed in the state where the preprocessing is completed and the image pool 23a is formed in the server device 20.

[0056] First, the installation image is captured in the recorder device 10 (step S201). In step S201, similarly to step S101, the captured installation image is sent to the server device 20 via the network N. In the server device 20, the communication unit 21 receives the installation image. That is, the communication unit 21 functions as an image acquisition unit that acquires the captured image captured by the imaging unit 11 installed in the cabin 104 of the construction machine.

[0057] Next, the processing unit 22 of the server device 20 generates a feature point image from the installation image (step S202). Step S202 also generates an image of the window frame portion extracted from the installation image as a feature point image, similar to step 102. That is, the processing unit 22 functions as a feature point extraction unit that extracts feature points of the vehicle interior 104 from the captured image.

[0058] Next, the processing unit 22 compares the feature point image generated in step S202 with the image group accumulated in the image pool 23a (step S203). Then, as a result of comparing the feature point image with the image group accumulated in the image pool 23a, the processing unit 22 determines whether there is a consistent image (step S204). That is, the processing unit 22 functions as an installation state determination unit that determines whether the installation state of the imaging unit 11 is appropriate based on the image pool 23a (pre-prepared information related to feature points for each type and model of construction machinery) and the extracted feature point image.

[0059] If there is no matching image (step S204; No), the processing unit 22 regards the category or model as different and issues a notification urging correct installation (step S205). That is, the processing unit 22 determines the installation status as "No". The image pool 23a stores feature point images of the category (backhoe, bulldozer, etc.) and model (XX, etc., manufactured by ○○ Company). The absence of a matching image in the image pool 23a means that it can be determined that the recorder device 10 is not installed on the corresponding construction machine, so the correct installation is urged.

[0060] As a method of urging correct settings, possible improper settings information indicating that the settings are incorrect may be sent to the recorder device 10 via the communication unit 21, and notification may be made in the recorder device 10 through display or sound output to urge correct settings, or possible improper settings information may be sent to a pre-registered email address, SNS, etc. via the communication unit 21. That is, the communication unit 21 functions as an installation notification unit that notifies when the installation status determination unit determines that the installation status is "No".

[0061] In step S205, not only the case of different types or models but also the case of not being installed in the correct position is included. For example, when it is instructed to be installed at the upper left rear portion of the seat, if it is installed at a different position, the feature point image is different, and therefore it is inconsistent with the image group accumulated in the image pool 23a. In such a case, it can also be determined that it is not correctly installed, and the correct installation can be urged.

[0062] On the other hand, if there is a matching image (step S204; Yes), the processing unit 22 determines whether it matches the last installation image (step S206). The last installation image is an image captured and stored when the recorder device 10 was activated and step S201 was executed last time.

[0063] If the image is inconsistent with the last installation image (step S206; No), the processing unit 22 considers that a replacement has occurred and issues a notification to urge the correct setting (step S207). That is, the acquisition of an image different from the last startup is considered to be likely to be replaced, for example, for the purpose of recycling the recorder device 10 in multiple construction machines. When a replacement occurs, the adhesive force of the double-sided tape of the mounting parts of the recorder device 10 decreases, which may cause it to fall, so the correct setting is urged. That is, the communication unit 21 functions as a replacement notification unit that notifies when the replacement determination unit determines that a replacement has occurred.

[0064] As a method of urging correct settings, possible bad settings information indicating that the settings are incorrect may be sent to the recorder device 10 via the communication unit 21 as in step S205, and notification may be made in the recorder device 10 through display or sound output to urge correct settings, or the possible bad settings information may be sent to a pre-registered email address, SNS, etc. via the communication unit 21. In addition, the possible bad settings information is preferably different from the content indicating that a change has occurred in step S205.

[0065] On the other hand, when the images are consistent (step S206; yes), the processing unit 22 determines the behavior inference library (step S208). The behavior inference library is a parameter set of a program that can infer the action of the construction machinery based on the data (acceleration data, etc.) obtained by the recorder device 10. That is, the behavior inference library is used to determine the action of the construction machinery, and can infer actions such as moving forward and digging and obtain information related to the action other than the image. The behavior inference library is set according to each category of construction machinery such as backhoe excavators and bulldozers. Therefore, the behavior inference library corresponding to the category represented by the image of the image pool that is consistent in step S204 is determined.

[0066] In addition, the server device 20 may at least perform the process up to the determination of the behavior estimation library, and analysis using the behavior estimation library may be performed by other servers or the like.

[0067] In addition, in step S208, the installation image obtained in step S201 is saved (recorded) in the storage unit 23 or the like as the last installation image used in step S206. That is, the image executed up to step S208 is an image in which the type or model is not different or replaced, and it can be said that the image is correctly installed, so it is saved as the reference image for the next startup. That is, the processing unit 22 functions as a feature point image recording unit that records the feature points when the installation state is determined to be "appropriate" as a feature point image.

[0068] As can be seen from the above description, steps S204 and S206 determine whether the installation state of the imaging unit 11 is appropriate. Then, through steps S204 and S206, the processing unit 22 functions as a category discrimination unit that discriminates the category of the engineering machinery based on the image pool 23a (pre-prepared information related to the feature points of each category of the engineering machinery) and the extracted feature point image. Then, through step S208, the processing unit 22 functions as a determination unit that determines the behavior estimation library (action parameter) for the imaging unit 11 corresponding to the determined category.

[0069] according to Figure 5 In the flowchart of FIG. 1 , step S201 functions as an image acquisition process, step S202 functions as a feature point extraction process, steps S204 and S206 function as a category determination process, and step S208 functions as a determination process. In addition, steps S204 and S206 also function as an installation state determination process.

[0070] According to the present embodiment, the server device 20 obtains the installation image captured by the camera unit 11 installed in the cabin 104 of the construction machinery through the recorder device 10, and extracts the feature point image in the cabin 104 from the installation image through the processing unit 22. Then, the processing unit 22 determines the category of the construction machinery based on the pre-prepared image pool 23a and the feature point image, and determines the behavior inference library for the recorder device 10 corresponding to the determined category. Thus, the category of the construction machinery can be automatically determined based on the image captured by the camera unit 11, and an appropriate behavior inference library can be determined. Therefore, even when the model is made common, the operation based on the setting and the like will not become complicated. In addition, the behavior inference library corresponding to the correct category of the construction machinery can be used, so that the reduction of the accuracy of the behavior inference can be prevented.

[0071] In addition, the characteristic point is the window frame of the vehicle interior 104. The window frame of the vehicle interior 104 is a part that greatly differs depending on the type of construction machinery, and when determining the type, it is possible to make a more accurate determination.

[0072] In addition, the images accumulated in the image pool 23a are not limited to the category unit, but are subdivided into each model. It is possible to specify each manufacturer, each model, etc., and the behavior estimation library can be subdivided compared with the category.

[0073] In addition, the server device 20 obtains the installation image captured by the imaging unit 11 installed in the cabin 104 of the construction machine through the recorder device 10, and extracts the feature point image in the cabin 104 from the installation image through the processing unit 22. Then, the processing unit 22 determines whether the installation state of the imaging unit 11 is appropriate based on the image pool 23a and the feature point image prepared in advance. In this way, it is possible to determine whether the imaging unit 11 is correctly installed based on the image captured by the imaging unit 11.

[0074] Furthermore, when the processing unit 22 determines that the installation state is "No", information indicating the notification is transmitted via the communication unit 21. Thus, when the installation state is inappropriate, the user or the like can be notified and urged to install correctly.

[0075] In addition, feature points when the processing unit 22 determines that the mounting state is "appropriate" are saved as feature point images, and when feature points extracted from the newly acquired mounting image are inconsistent with the saved feature point images, it is determined that the recorder device 10 has been replaced. In this way, it is possible to detect the replacement of the recorder device 10 based on the mounting image. For example, it is possible to prevent the adhesive force of the double-sided tape of the mounting component from deteriorating due to replacement, thereby preventing the camera from falling off.

[0076] Furthermore, when the processing unit 22 determines that replacement has occurred, information indicating the replacement is transmitted via the communication unit 21. This can prompt the user to replace the double-sided tape or the mounting member in order to prevent the imaging unit from falling.

[0077] In addition, with respect to the above-mentioned last installation image, an image further subdivided according to each installation angle of the recorder device 10 (shooting unit 11) can also be generated. The recorder device 10 (shooting unit 11) is not limited to the standard installation angle, and sometimes changes are allowed as long as it is within the specified angle range. Therefore, for example, when it is determined in step S208 to be used as the last installation image, a plurality of images in which the installation angle is changed within the allowable range are generated, and the plurality of images are processed as the last installation image (group). Thus, in the case where the angle is adjusted until the next time, the installation status is not judged as "no" if it is not replaced. Therefore, a judgment that is more in line with the actual situation can be made.

[0078] In addition, the images segmented for each installation angle are not limited to replacement determination, but can also be used when distinguishing the type and model of the construction machinery in step S204. By using them when distinguishing the type and model of the construction machinery, it is possible to prevent erroneous determination that the type or model is different when the angle is changed.

[0079] The above-mentioned images segmented for each mounting angle are preferably automatically generated from the mounting image by the processing unit 22. Automatic generation eliminates the need to actually change the mounting angle to acquire the image, which can save the user's time and effort.

[0080] Alternatively, feature point images segmented for each angle may be stored in the image pool 23a. This allows feature point images to be generated for each angle each time, without the need to generate images within the angle range each time from the last installation image.

[0081] Furthermore, the above embodiment is a structure in which the discrimination device and the determination device coexist, but it may also be a structure having only one function. Here, in the case of a discrimination device having an image acquisition unit, a feature point extraction unit, a category discrimination unit, and a determination unit, the information related to the feature points may be information related to the feature points of each category of construction machinery, and may not be subdivided for each machine type.

[0082] In addition, in the above-mentioned structure, the image captured by the recorder device 10 is sent to the server device 20, and the server device 20 performs the classification and the determination of the appropriateness of the installation state, but the classification and the determination of the appropriateness of the installation state may be performed by, for example, the control unit 13 of the recorder device 10. In this case, the image pool 23a may be downloaded in advance via the network, or may be obtained from a memory card or the like. By adopting a structure in which the classification and the determination of the appropriateness of the installation state are performed by the recorder device 10, the classification and the determination can be performed even in an environment in which the network cannot be connected.

[0083] In the above-mentioned structure, the image pool 23a is generated in the server device 20, and the category is determined and the appropriateness of the installation state is determined. However, the generation of the image pool 23a may be performed in other server devices, etc. That is, an information processing device having a feature point acquisition unit for acquiring feature points and an accumulation unit for accumulating the acquired feature points may be provided. Thus, for example, by accumulating feature points in the server device, etc., it is possible to easily cope with new models, etc.

[0084] In addition, the present invention is not limited to the above-described embodiments. That is, those skilled in the art can carry out various deformations and implement them according to the previously known views without departing from the scope of the gist of the present invention. According to this deformation, as long as there is a discrimination device or a determination device of the present invention, it is certainly also included in the scope of the present invention.

[0085] Description of symbols

[0086] 10: Recorder device; 11: Photographing unit; 15: Recording unit; 20: Server device (discriminating device, determining device); 21: Communication unit (image acquiring unit, feature point acquiring unit, installation notification unit, replacement notification unit); 22: Processing unit (feature point extraction unit, category discriminating unit, determining unit, installation status determining unit, feature point image recording unit, replacement determining unit); 23: Storage unit (accumulation unit); 23a: Image pool (information related to feature points of each category of construction machinery); 100: Construction machinery; 104: Vehicle cabin.

Claims

1. A discrimination device, characterized in that: The identification device has: an image acquisition unit that acquires an image captured by an image capture unit installed in a cabin of the construction machine; a feature point extraction unit for extracting feature points inside and outside the vehicle interior from the captured image; a category determination unit that determines the category of the construction machine based on information related to the feature points of each category of the construction machine prepared in advance and the extracted feature points; as well as A determination unit determines an operation parameter for the imaging unit corresponding to the determined type.

2. The identification device according to claim 1, characterized in that: The information related to the feature point is a feature point provided on a seat in the vehicle interior or a window frame in the vehicle interior.

3. The identification device according to claim 1, characterized in that: The information related to the characteristic points is further subdivided for each model of the construction machinery in the category.

4. The identification device according to claim 1, characterized in that: The motion parameters are used to determine the motion of the engineering machine.

5. A photographing device, characterized in that: The camera has: The identification device according to any one of claims 1 to 4; the photographing unit; and A recording unit records the captured image acquired by the image acquisition unit.

6. An information processing device, characterized in that: The information processing device has: a feature point acquisition unit that acquires the feature points extracted by the determination device according to any one of claims 1 to 4; and An accumulation unit accumulates the acquired feature points.

7. A method for distinguishing a type of construction machinery, the method being performed by a distinguishing device for distinguishing a type of construction machinery, characterized in that: The identification method includes the following steps: An image acquisition step of acquiring an image captured by a camera installed in a cabin of the construction machine; A feature point extraction step of extracting feature points inside and outside the vehicle interior from the captured image; a category determination step of determining the category of the construction machinery based on information related to the feature points of each category of the construction machinery prepared in advance and the extracted feature points; as well as The step of determining an operation parameter for the imaging unit corresponding to the determined category.

8. A discrimination procedure, characterized in that: The determination program causes a computer to execute the determination method according to claim 7 .

9. A recording medium capable of being read by a computer, characterized in that: The recording medium stores the determination program according to claim 8.

Images