JP2019190805A - Information processing device, control device and improper object detection system - Google Patents

Abstract

To further readily provide detection of an incineration improper object more than prior art.SOLUTION: An information processing device (1) includes: a trash image acquisition unit (101) for acquiring a trash image filming trash in the middle of dropping into a refuse pit following transported to a refuse incineration facility; and an improper object detection unit (104) for detecting an incineration improper object from the trash image.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information processing apparatus and the like that detect inadequate materials for incineration from garbage carried into a waste incineration facility.

  In recent years, illegal importation of unsuitable materials into garbage incineration facilities has become a problem. Inappropriate incinerators are thrown into the incinerator, resulting in worse combustion in the incinerator, clogging in the incinerator ash removal equipment, etc., and in some cases, the incinerator may be stopped urgently .

  At present, the workers of the garbage incineration facility randomly select the collected garbage and inspect the selected garbage manually. For this reason, the burden of a worker is large.

  In Patent Document 1 below, as a technique for solving this problem, an inspection room in which a garbage truck can be placed is provided, and X-rays are irradiated to the entire garbage collection part of the garbage truck in the inspection room, so that it is not suitable for incineration. A technique for detecting an object is disclosed.

JP 08-166359 A (published June 25, 1996)

  However, the conventional technology as described above has a problem that it is difficult to realize because it is necessary to provide an inspection room in a garbage incineration facility in order to detect inadequate incineration.

  One embodiment of the present invention has been made in view of such problems, and an object of the present invention is to more easily realize detection of incinerators in comparison with the prior art.

  In order to solve the above-described problem, an information processing apparatus according to an aspect of the present invention includes a waste image acquisition unit that acquires a waste image that is taken into a waste incineration facility and captures waste that is being dropped into a waste pit. And a detection unit for detecting an incineration inappropriate material from the garbage image.

  In order to solve the above-described problem, an inappropriate object detection system according to one aspect of the present invention includes an imaging device that captures an image of waste that has been carried into a waste incineration facility being dropped into a waste pit, and an information processing device. The information processing apparatus is configured to acquire a garbage image obtained by photographing the garbage from the photographing apparatus and detect an incineration inappropriate substance from the garbage image.

  According to one aspect of the present invention, there is an effect that it is possible to more easily realize detection of an incineration-inappropriate object as compared with the prior art.

It is a block diagram which shows an example of the principal part structure of the information processing apparatus which concerns on one Embodiment of this invention. It is a block diagram which shows an example of a principal part structure of the unsuitable thing detection system which concerns on one Embodiment of this invention. It is a figure which shows a mode that a garbage truck is dropping garbage into a garbage pit in a garbage incineration facility. It is a figure which shows the inside of a garbage pit. (A) is a figure which shows an example of the garbage image image | photographed with the garbage imaging device shown in FIG. 2, (b) is a figure which shows an example of the vehicle image image | photographed with the vehicle imaging device shown in FIG. is there. 3 is a flowchart illustrating an example of a flow of processing executed by the information processing apparatus illustrated in FIG. 1. It is a figure which shows one specific example of the extraction process of a trash area. It is a figure which shows one specific example of the extraction process of a trash area. It is a figure which shows the example of classification | category of incineration inappropriate material according to a risk. It is a figure which shows one specific example of the threshold value information memorize | stored in the memory | storage part of the information processing apparatus shown in FIG. 3 is a flowchart illustrating an example of a flow of processing executed by the information processing apparatus illustrated in FIG. 1. It is a figure which shows one specific example of the search process of the whole garbage image. It is a figure which shows an example of the matching table memorize | stored in the memory | storage part of the information processing apparatus shown in FIG. It is a figure which shows the structural example of the garbage storage part provided in the front | former stage of the garbage pit. It is a block diagram which shows an example of a principal part structure of the information processing apparatus which detects an incineration inappropriate material based on a sound or vibration.

Embodiment 1
An information processing apparatus 1 according to an embodiment of the present invention will be described with reference to FIGS. The information processing apparatus 1 has a function of detecting unsuitable materials for incineration from waste carried into a waste incineration facility. Specifically, the information processing device 1 detects an incineration inappropriate object using an image obtained by photographing the garbage being dropped into the garbage pit. Here, an incineration inappropriate material is an object which should not be incinerated in the incinerator provided in the garbage incineration facility. Specific examples of incineration inappropriate materials will be described later.

〔System configuration〕
A configuration of a system including the information processing apparatus 1 will be described with reference to FIG. FIG. 2 is a block diagram illustrating an example of a main configuration of the inappropriate object detection system 100 that is an example of the system. The inappropriate object detection system 100 includes an information processing device 1 (control device), a garbage photographing device 2 (photographing device), a vehicle photographing device 3, an alarm device 4, a display device 5, a data server 6, a printing device 7, and a crane 8. Contains. The information processing apparatus 1 is communicably connected to other apparatuses included in the inappropriate object detection system 100, and transmits / receives data or transmits control signals. Since the printing apparatus 7 will be described in detail in a second embodiment to be described later, description thereof is omitted here.

  The garbage photographing device 2 photographs garbage on the way to the garbage pit, and transmits the photographed image to the information processing device 1. Hereinafter, this captured image is referred to as a garbage image. The vehicle photographing device 3 takes in garbage, photographs a vehicle (so-called garbage truck) that drops the garbage into the garbage pit, and transmits the photographed image to the information processing device 1. Hereinafter, this captured image is referred to as a vehicle image. As an example, the garbage photographing device 2 may be a high-speed shutter camera that photographs a moving image. The vehicle photographing device 3 is not particularly limited as long as it is a device having a photographing function.

  FIG. 3 is a diagram showing a state where a garbage truck is dropping garbage into a garbage pit in a garbage incineration facility. FIG. 4 is a diagram showing the inside of the garbage pit. As shown in FIG. 3, the waste incineration facility is provided with a plurality of doors such as doors 300 </ b> A and 300 </ b> B (hereinafter collectively referred to as door 300 when it is not necessary to distinguish them). In addition, as shown in FIG. 4, a garbage pit is provided at the end of the door 300. That is, when the door 300 is opened, a drop opening for dropping garbage into the garbage pit appears. As shown in FIG. 3, the garbage truck 200 drops garbage into a garbage pit from a drop opening.

  In order to capture the garbage image and the vehicle image, the garbage photographing device 2 and the vehicle photographing device 3 may be attached to the attachment portions 400 of the doors 300 as shown in FIGS. 3 and 4 as an example. Specifically, the dust photographing apparatus 2 may be attached to the attachment portion 400 in a direction and an angle in which the region 600 in FIG. 4 can be photographed. Further, the vehicle photographing device 3 may be attached to the attachment location 400 at a direction and an angle in which the garbage truck 200 can be photographed. The attachment location 400 is an example of the attachment location of the garbage imaging device 2 and the vehicle imaging device 3, and the installation location of the garbage imaging device 2 is an installation location 400 as long as it can capture the region 600 of FIG. It may be a different location. Moreover, the attachment location of the vehicle photographing device 3 may be a location different from the attachment location 400 as long as the garbage collection vehicle 200 can be photographed.

  5A is a diagram illustrating an example of a garbage image photographed by the garbage photographing device 2, and FIG. 5B is a diagram illustrating an example of a vehicle image photographed by the vehicle photographing device 3. is there. As shown in FIG. 5 (a), when garbage is introduced into the garbage pit from the garbage truck 200 by attaching the garbage photographing device 2, the garbage on the way to the garbage pit is reflected. A garbage image is taken. Further, by attaching the vehicle photographing device 3 as described above, a vehicle image in which the license plate 900 of the garbage truck 200 is photographed is photographed as shown in FIG. By including the license plate 900 in the vehicle image, the facility detection system 100 or a facility incineration facility that has seen the vehicle image can identify the garbage truck 200 that appears in the vehicle image. In addition, the vehicle image should just contain the information which can identify the garbage truck 200, and this information is not limited to the license plate 900. Examples of the information include the shape and color of the vehicle, and the name (for example, company name) and identification number of the organization to which the importer belongs, described on the vehicle.

  The garbage photographing device 2 and the vehicle photographing device 3 may have a function of starting photographing when it is detected that the garbage truck 200 has come in front of the door 300. As a result, it is possible to prevent photographing when dust is not dropped. In other words, it is possible to reduce images that are unnecessary for detection of unsuitable materials for incineration. In addition, the garbage photographing device 2 starts photographing when it detects that the door 300 is opened instead of the function of starting photographing when it is detected that the garbage truck 200 has come in front of the door 300. It may have a function. Thereby, the image unnecessary for the detection of an unsuitable thing for incineration can be reduced more.

  In addition, as an example, when the garbage collection vehicle 200 enters the garbage incinerator from which door 300 the garbage is dropped into the garbage pit, it is determined by the facility personnel of the garbage incineration facility. . Then, the determined door 300 is opened, for example, when a facility staff operates a predetermined device. The facility staff may ask the person who carries in the garbage incineration facility with the garbage truck 200 to present information for identifying the carrying subject when entering the waste incineration facility. In addition, a carrying-in main body refers to a carrying-in person or the group (for example, collection company etc. which collects garbage) to which this carrying-in belongs. Further, the information for specifying the import subject may be, for example, the name of the import subject or a number described on the license plate 900 of the garbage truck 200.

  Referring again to FIG. 2, the description returns to the main configuration of the inappropriate object detection system 100. The alarm device 4 outputs an alarm under the control of the information processing device 1 when an incineration inappropriate material is detected. Note that the type of alarm is not particularly limited as long as the importer can recognize that an incineration inappropriate material has been detected. For example, the alarm may be output by turning on or blinking the light emitting device or outputting a predetermined sound. The alarm device 4 is preferably provided for each drop port. As an example, the light 500 illustrated in FIG. 3 may also serve as the alarm device 4. In this example, the alarm output may be blinking of the light 500.

  The display device 5 displays the garbage image photographed by the garbage photographing device 2 when the garbage is dropped. The display device 5 may be provided on each drop opening (that is, the door 300) as an example. Thereby, the display device 5 can show the waste image to the importer, and can give the importer a consciousness that “the waste is being dropped”. As a result, it is possible to prevent the incineration inappropriate material from being carried in.

  Further, the display device 5 can confirm whether or not an incineration inappropriate object has been dropped by looking at a garbage image displayed by a facility member in a monitoring room (not shown) provided to face the door 300. It is preferable to display the garbage image with a size of about.

  The data server 6 stores data acquired from the information processing apparatus 1. As an example, the data server 6 stores garbage images and vehicle images.

  As shown in FIG. 4, the crane 8 is provided in the garbage pit, and carries the garbage in the garbage pit under the control of the information processing apparatus 1.

〔Device configuration〕
The configuration of the information processing apparatus 1 will be described with reference to FIG. FIG. 1 is a block diagram illustrating an example of a main configuration of the information processing apparatus 1. The information processing apparatus 1 according to the present embodiment has a function of detecting an unsuitable object for incineration using the garbage image when the garbage image is acquired from the garbage photographing apparatus 2. That is, the information processing apparatus 1 according to the present embodiment detects an unsuitable incinerator while the garbage truck 200 is dropping the garbage into the garbage pit or immediately after the dumping. As a result, the information processing apparatus 1 can notify the importer and the facility staff that the incinerator inappropriate object has been detected before the importer who dropped the incinerated object into the garbage pit leaves the waste incineration facility. it can. The information processing apparatus 1 may be a personal computer as an example.

  The information processing apparatus 1 includes a control unit 10, a storage unit 11, and a communication unit 12. The control unit 10 controls each unit of the information processing apparatus 1 in an integrated manner. The storage unit 11 stores various data used by the information processing apparatus 1. The communication unit 12 receives data from each device included in the inappropriate object detection system 100, and transmits data to each device. The storage unit 11 may be provided outside the information processing apparatus 1 as with the data server 6, or the data server 6 may function as the storage unit 11.

  The control unit 10 includes a dust image acquisition unit 101, an image region extraction unit 102 (extraction unit), a dust identification unit 103 (identification unit), an inappropriate object detection unit 104 (detection unit), an alarm control unit 105, and a display control unit 106. , A crane control unit 107, a vehicle image acquisition unit 108, and an association unit 109 are included. Also. The storage unit 11 includes at least threshold information 111, carry-in subject identification information 112, and an association table 113. Note that details of the associating unit 109, the carry-in subject specifying information 112, and the associating table 113 will be described in the second embodiment to be described later, and the description thereof is omitted here.

  The garbage image acquisition unit 101 acquires a garbage image from the garbage photographing device 2 via the communication unit 12. The garbage image acquisition unit 101 outputs the acquired garbage image to the image region extraction unit 102. In the present embodiment, the garbage image obtaining unit 101 obtains a moving image composed of a plurality of frame images photographed by a high speed shutter camera that is the garbage photographing device 2. In other words, the garbage image acquisition unit 101 acquires a plurality of garbage images in time series until the dropped garbage falls into the garbage pit.

  The garbage image acquired by the garbage image acquisition unit 101 is provided with date / time information indicating the date and time when the garbage image was taken and door identification information for identifying the door 300 to which the garbage photographing apparatus 2 that photographed the garbage image is attached. Has been. In other words, the door identification information can also be expressed as information for identifying the drop-out port into which the garbage shown in the garbage image has been thrown.

  The garbage image acquisition unit 101 transmits the acquired garbage image to the data server 6 via the communication unit 12. That is, the garbage image acquisition unit 101 stores the acquired garbage image in the data server 6. In addition, the garbage image acquisition unit 101 outputs the acquired garbage image to the display control unit 106.

  The image area extraction unit 102 extracts a specific image area from the garbage image. Specifically, the image region extraction unit 102 extracts a plurality of garbage images, that is, image regions changing in time series from each frame image of the acquired moving image. As described above, since the garbage image is an image taken in the middle of dropping the garbage into the garbage pit, the image area changing in time series is an area where the garbage is reflected. That is, the image area extraction unit 102 extracts an area in which garbage appears from the garbage image. The image area extraction unit 102 adds information indicating the extracted image area (hereinafter referred to as a garbage area) to the acquired garbage image. Then, the image area extraction unit 102 outputs the garbage image to the garbage identification unit 103 together with the assigned date and time information and door identification information. Details of the dust region extraction processing by the image region extraction unit 102 will be described later.

  The garbage identifying unit 103 identifies the type of garbage that appears in the garbage image. Specifically, the garbage identification unit 103 identifies the type of garbage in the garbage area based on the garbage image to which the information indicating the garbage area is given, acquired from the image area extraction unit 102. This identification function can be realized by an object recognition algorithm (for example, using a neural network) that recognizes an object in an image. When such an algorithm is used, machine learning is performed in advance using a training data set. Although the algorithm to be used is not particularly limited, it is preferable to use a deep learning algorithm (AlexNet, GoogLeNet, ResNet, etc.) from the viewpoint of determination accuracy and the like.

  The garbage identification unit 103 gives date / time information and door identification information to the identification result, and outputs the result to the unsuitable object detection unit 104.

(Machine learning)
Here, the details of the above-described machine learning will be described. In order for the garbage identification unit 103 to correctly identify the garbage in the garbage image, the garbage identification unit 103 performs machine learning in advance. The training data set described above includes a training image in which an object is captured. Each training image is associated with data indicating a correct determination result based on the training image. Specifically, the training image in which the garbage bag is shown is associated with data indicating the region in which the garbage bag is shown and data indicating that the object in the region is a garbage bag, and the training data is It is configured.

  The training data set is a set including a plurality of such training data. Training data is prepared for each class set to identify the type of garbage. For example, when the types of garbage are classified into one of seven classes, the number of training images corresponding to each class is prepared so as to obtain a desired determination system (for example, 100 for each class). The seven classes are, for example, as follows. (1) Sphygmomanometer, (2) Thermometer, (3) Automotive battery, (4) Clothesline, (5) Rock, (6) Ceramic, (7) Transparent garbage bag.

  In addition, it is preferable that the training image is an image of an object taken out as garbage so that the garbage in the garbage image can be correctly identified. Specifically, the training image is preferably an image in which an object in which deformation, damage, contamination, corrosion, or the like has occurred is shown. Further, the number of training data may be increased by inverting, rotating, enlarging, or reducing the training image vertically and horizontally. Further, the class to be set is not limited to the above seven classes.

  In the machine learning, even if the training data is used for the training data to identify the type of garbage and update the parameters related to the identification according to the identification result, the training data may be processed. Good. Thereby, it is possible to improve the identification accuracy of the dust identification unit 103 and make the dust identification unit 103 in a state in which identification can be performed with an accuracy enough to withstand practical use.

  Returning to the description of the configuration of the information processing apparatus 1 with reference to FIG. The unsuitable object detection unit 104 detects an unsuitable object for incineration from the garbage image. Specifically, the inappropriate material detection unit 104 determines whether or not the type of garbage indicated by the identification result is an incineration inappropriate material based on the identification result acquired from the waste identification unit 103 and the threshold information 111. For example, it is determined whether or not an incineration inappropriate material has been dropped. The unsuitable object detection unit 104 gives the door identification information given to the identification result to the result of this determination (hereinafter referred to as the detection result), and outputs it to the alarm control unit 105 and the crane control unit 107. The details of the threshold information 111 and the details of the incompatibility detection processing by the inadequate matter detection unit 104 will be described later.

  The vehicle image acquisition unit 108 acquires a vehicle image from the vehicle imaging device 3 via the communication unit 12. The vehicle image acquisition unit 108 transmits the acquired vehicle image to the data server 6 via the communication unit 12. That is, the vehicle image acquisition unit 108 stores the acquired vehicle image in the data server 6. In the present embodiment, the vehicle image acquisition unit 108 acquires an image captured by a camera that is the vehicle imaging device 3. The image is an image obtained by photographing the garbage truck 200 existing in front of the door 300 (that is, the outlet). The vehicle image capturing timing is not particularly limited as long as it is within a period in which the garbage truck 200 exists in front of the door 300.

  Note that the vehicle image acquired by the vehicle image acquisition unit 108 includes date information indicating the date and time when the vehicle was captured, and the vehicle imaging device 3 that captured the vehicle image, as with the garbage image acquired by the garbage image acquisition unit 101. Door identification information for identifying the attached door 300 is given.

  The alarm control unit 105 controls the alarm device 4 to output an alarm when the incineration inappropriate object is dropped. Specifically, when the detection result acquired from the inappropriate material detection unit 104 indicates that the incineration inappropriate material has been dropped, the alarm control unit 105 is given to the detection result via the communication unit 12. A control signal is transmitted to the alarm device 4 specified from the door identification information. Thereby, the alarm device 4 outputs an alarm.

  The display control unit 106 controls the display device 5 to display a garbage image. Specifically, when the display control unit 106 acquires a garbage image from the garbage image acquisition unit 101, the display control unit 106 transmits the garbage image and the control signal from the door identification information associated with the garbage image via the communication unit 12. It transmits to the specified display device 5. Thereby, the display device 5 displays the garbage image.

  The crane control unit 107 controls the operation of the crane. As an example, the crane control unit 107 moves an unsuitable incinerator that has fallen into the garbage pit to a position that can be distinguished from incineratorable waste when dropping of the unsuitable incinerator is detected. Specifically, when the detection result acquired from the unsuitable object detection unit 104 indicates that the unsuitable incinerator has been dropped, the crane control unit 107 does not properly incinerate the crane 8 via the communication unit 12. A control signal is transmitted to move the object to a position that can be distinguished from incinerated garbage. In addition, the detail of the movement process of incineration inappropriate material by the crane control part 107 is mentioned later.

[Process flow]
Based on FIG. 6, a flow of processing executed by the information processing apparatus 1 will be described. FIG. 6 is a flowchart illustrating an example of a flow of processing executed by the information processing apparatus 1.

  The garbage image acquisition unit 101 acquires a garbage image from the garbage photographing apparatus 2 (S101). Moreover, the vehicle image acquisition part 108 acquires a vehicle image from the vehicle imaging device 3 (S102). Then, the garbage image acquisition unit 101 and the vehicle image acquisition unit 108 store the acquired garbage image and vehicle image in the data server 6 (S103). In addition, the garbage image acquisition unit 101 outputs the acquired garbage image to the display control unit 106. The display control unit 106 displays the acquired garbage image on the display device 5 (S104).

  In addition, the garbage image acquisition unit 101 outputs the acquired garbage image to the image region extraction unit 102. The image area extraction unit 102 compares a plurality of garbage images acquired in succession, and determines whether or not there is an image area changing in time series in the garbage image (S105). When it is determined that there is an image area changing in time series (YES in S105), the image area extraction unit 102 extracts the image area as a garbage area (S106), and acquires information indicating the garbage area Applies to garbage images. Then, the image area extraction unit 102 outputs a garbage image provided with information indicating the garbage area to the garbage identification unit 103. On the other hand, if it is determined that there is no image area changing in time series (NO in S105), the process shown in FIG. 6 returns to step S101.

(Specific example of garbage area extraction processing)
Here, a specific example of the dust region extraction processing in step S106 will be described with reference to FIGS. 7 and 8 are diagrams showing a specific example of the dust region extraction process.

  The trash images 21 to 23 illustrated in FIG. 7 are a plurality of trash images acquired by the image region extraction unit 102 in succession. As shown in FIG. 7, the garbage image 21 is a garbage image taken at time t−Δt, the garbage image 22 is a garbage image taken at time t, and the garbage image 23 is garbage taken at time t + Δt. It is an image. That is, the garbage images 21 to 23 are taken in the order of the garbage images 21, 22, and 23. In the garbage images 21 to 23, the object 211 dropped from the garbage truck 200 moves toward the garbage pit along the route provided between the dumping opening and the garbage pit as time elapses ( (Sliding down) is shown.

  As an example, the image area extraction unit 102 specifies a garbage area, that is, an image area changing in time series from a plurality of acquired garbage images using an inter-frame difference method. Here, an example in which a dust region is extracted from the image 22 will be described. Specifically, the image area extraction unit 102 first performs gray scale conversion on the garbage image 21 and the garbage image 22. Then, the image area extraction unit 102 calculates a difference value for each pixel between the trash image 21 and the trash image 22 subjected to gray scale conversion, and obtains a difference value matrix. The image area extraction unit 102 performs processing for the difference value matrix so that the value is white if the value of each element is equal to or greater than a predetermined value, and is black if the value is less than the predetermined value. As a result, a binarized image 24 between the garbage image 21 and the garbage image 22 is obtained in which the changed area is white and the unchanged area is black. In the binarized image 24, an area 212A corresponding to the object 211 in the garbage image 21 and an area 211B corresponding to the object 211 in the garbage image 22 are white.

  The image region extraction unit 102 performs a similar process between the garbage image 22 and the garbage image 23 to obtain a binarized image 25. In the binarized image 24, the region 212B and the region 211C corresponding to the object 211 in the garbage image 23 are white.

  The image area extraction unit 102 performs AND processing on the obtained binarized images 24 and 25. As a result, the image area extraction unit 102 obtains a binarized image 26 in which only white areas in both the binarized images 24 and 25 are white as shown in FIG. In other words, the image area extraction unit 102 obtains the binarized image 26 including only the area 211B among the areas 211A to 211C.

  The image area extraction unit 102 surrounds each white area such as the area 211B in the binarized image 26 with a rectangle. Then, it is determined whether or not the rectangular area is larger than a predetermined threshold. When it is determined that the value is larger than the predetermined threshold, the image region extraction unit 102 determines that the garbage image 22 corresponding to the binarized image 26 is a garbage image when the garbage is dropped. Then, the image area extraction unit 102 arranges a rectangle 221 having the same area as the rectangle surrounding the area 211B at the position of the garbage image 22 corresponding to the position of the rectangle surrounding the area 211B in the binarized image 26. . As a result, as shown in FIG. 8B, a garbage image 27 in which most of the object 211 is surrounded by the rectangle 221 is obtained. The image area extraction unit 102 extracts an area surrounded by the rectangle 221 in the image 27 as a garbage area. The image area extraction unit 102 performs the above-described processing on the garbage images 21 and 23 and all other garbage images not shown. Thereby, in all the garbage images, it is determined whether or not it is a garbage image when the garbage is dropped, and a garbage region is extracted from the garbage image when the garbage is dropped.

  Note that a method for identifying an image region changing in time series from a plurality of acquired garbage images is not limited to the inter-frame difference method. As an example, the image region extraction unit 102 may use an optical flow.

  7 and 8 show an example in which one object 211 is shown in the image, there may be a case where a plurality of objects are shown in the image. In this case, in accordance with the above-described method described with reference to FIGS. 7 and 8, for each of the plurality of objects, a region surrounded by a rectangle is extracted as a dust region.

  Again, referring back to FIG. 6, the description returns to the process flow. The garbage identifying unit 103 identifies the type of garbage from the garbage area of the acquired garbage image (S107). As an example, the dust identifying unit 103 performs object detection and outputs scores of each set class (for example, the above seven classes) for the objects in the dust region. The output score is, for example, a numerical value with two decimal places in the range from 0 to 1, and the higher the score, the higher the possibility of being an object of each class. The garbage identification unit 103 outputs the output seven scores to the unsuitable object detection unit 104 as identification results. Note that the dust identifying unit 103 may output only the highest score among the seven scores to the unsuitable object detecting unit 104.

  The unsuitable object detection unit 104 determines whether the identified garbage is an unsuitable object for incineration (S108). Then, the unsuitable object detection unit 104 outputs the determination result (detection result) to the alarm control unit 105 and the crane control unit 107.

(Specific example of incineration inappropriate material detection processing)
Here, a specific example of the incineration inappropriate material detection process in step S108 will be described with reference to FIG. 9 and FIG. FIG. 9 is a diagram illustrating a classification example of incineration inappropriate materials according to the degree of risk. FIG. 10 is a diagram illustrating a specific example of the threshold information 111 stored in the storage unit 11. Here, an example in which the above seven classes are set in object detection will be described.

  As shown in FIG. 9, incineration inappropriate materials can be classified into one of three risk levels of high, medium, and low as an example. Inappropriate incinerators with a high risk level are inadequate incinerators that may stop the entire incinerator. In the example of FIG. 9, a material containing a lot of heavy metals is regarded as an incineration inappropriate material with a high risk level. Examples of such incineration-inappropriate materials include high mercury content such as blood pressure monitors and thermometers, and high lead content such as automobile batteries. Inappropriate incinerators with a medium risk are incineration compounds that may shut down some facilities of the incinerator. In the example of FIG. 9, an item that may block the ash removal facility (conveyor) is regarded as an incineration inappropriate material with a risk of “medium”. Examples of such incineration-inappropriate materials include long objects such as clothespins, and heavy materials such as rocks. Inappropriate incinerators with a low risk are incinerators that worsen the combustion of incinerators, and generally include low-calorie objects (that is, incombustible waste). Examples include glass products and ceramics.

  In other words, the objects shown by the above seven classes are (1) sphygmomanometer, (2) thermometer, (3) automobile battery is an incineration inappropriate item with high risk, (4) clothespin, ( 5) The rock is an inappropriate incinerator with a medium risk, and (6) the ceramic is an incinerator with a low risk. On the other hand, (7) transparent garbage bags are not suitable for incineration.

  Based on the acquired identification result and threshold information 111, the unsuitable object detection unit 104 determines whether or not the dust in the garbage image corresponds to any one of the above three unsuitable incinerators. Specifically, as illustrated in FIG. 10, the threshold information 111 is information in which the threshold value of the score output by the garbage identification unit 103 is associated with the degree of risk. As shown in FIG. 10, the threshold information 111 is associated with a lower threshold as the degree of risk is higher. In other words, the threshold information 111 can detect garbage that has a high risk of being an inappropriate incinerator, even if the score is low, as being inadequate for incineration. Thereby, the stop of an incinerator can be prevented.

  The unsuitable object detection unit 104 compares the six scores in the classes (1) to (6) out of the seven scores with the corresponding risk threshold value. That is, the unsuitable object detection unit 104 sets (1) a sphygmomanometer, (2) a thermometer, and (3) a car battery score to a threshold (0.5 in the example of FIG. ). Further, the unsuitable object detection unit 104 compares the score of (4) clothespin and (5) rock with a threshold (0.7 in the example of FIG. 10) associated with the risk “medium”. Further, the unsuitable object detection unit 104 compares (6) the ceramic score with a threshold value (0.9 in the example of FIG. 10) associated with the risk level “low”. As a result of the comparison, if there is a score equal to or higher than the threshold value, the inappropriate material detection unit 104 determines that the identified garbage is an incineration inappropriate material. On the other hand, when there is no score equal to or higher than the threshold, the inappropriate material detection unit 104 determines that the identified garbage is not an incineration inappropriate material.

  Note that the above-described detection process for inadequate incineration is an example, and the detection process is not limited to this example. For example, the dust identifying unit 103 may identify the name of an object using an object detection method such as YOLO for the dust region, and output the name to the unsuitable object detecting unit 104 as the identification result. In the case of this example, the inappropriate object detection unit 104 is identified when the name of the incineration inappropriate object stored in the storage unit 11 instead of the threshold information 111 matches the name of the object acquired as the identification result. We judge that garbage is incineration unsuitable thing.

  Again, referring back to FIG. 6, the description returns to the process flow. If the inappropriate material detection unit 104 determines that the identified garbage is an incineration inappropriate material (YES in S108), the alarm control unit 105 that has acquired the detection result controls the alarm device 4 to output an alarm ( S109). Moreover, the crane control part 107 which acquired the detection result controls a crane, and moves an incineration inappropriate material (S110). On the other hand, when the inappropriate material detection unit 104 determines that the identified garbage is not an incineration inappropriate material (NO in S108), the alarm control unit 105 and the crane control unit 107 that have acquired the detection results respectively perform steps S109 and S110. Do not execute processing.

(Specific example of transfer processing of unsuitable for incineration)
Here, a specific example of the incineration inappropriate material moving process in step S110 will be described. As described above, the crane control unit 107 transmits, to the crane 8, a control signal for moving the unsuitable incinerator to a position where it can be distinguished from incinerated garbage. At this time, the crane control unit 107 identifies an area in the garbage pit where an incineration inappropriate material has been dropped based on the door identification information given to the detection result. Specifically, the information processing apparatus 1 associates the door identification information with the area information indicating the area of the garbage pit to which the garbage dropped from the drop-out port indicated by the door identification information is associated (not shown). ) Is stored in the storage unit 11 in advance. The crane control unit 107 refers to the information and identifies the region information associated with the acquired door identification information, thereby identifying the region where the incineration inappropriate material has been dropped. Then, the crane control unit 107 includes the area information indicating the identified area in the control signal, and transmits it to the crane 8. As a result, the crane 8 grasps the garbage in the area indicated by the area information, that is, the garbage containing the incineration-inappropriate material, and moves the caught garbage to an area predetermined as the area where the garbage not to be incinerated is arranged.

  Again, referring back to FIG. 6, the description returns to the process flow. When an operation to turn off the information processing apparatus 1 is input to an operation unit (not shown) of the information processing apparatus 1 (YES in step S111), the process illustrated in FIG. 6 ends. On the other hand, if the information processing apparatus 1 is not turned off (NO in step S111), the process shown in FIG. 6 returns to step S101.

[Embodiment 2]
An information processing apparatus 1 according to another embodiment of the present invention will be described with reference to FIGS. For convenience of explanation, members having the same functions as those described in the above embodiment are given the same reference numerals, and the description thereof will not be repeated.

  In the first embodiment, an example has been described in which the garbage collection vehicle 200 detects inadequate incineration during or after the dumping of the garbage into the garbage pit. Specifically, in the first embodiment, in order to detect incineration inappropriate materials in a short time, the image region extraction unit 102 identifies a region that changes in time series in the waste image as a waste region, It was determined whether or not the garbage area was unsuitable for incineration. In other words, in the first embodiment, since the entire garbage image is not searched, there is a possibility that an incineration inappropriate object cannot be detected without omission.

  On the other hand, this embodiment demonstrates the example which searches the whole garbage image and detects an incineration inappropriate material. In this example, since it takes time to detect the incineration inappropriate material, as in the example described in the first embodiment, the garbage collection vehicle 200 is dropping the garbage into the garbage pit or immediately after the dropping is finished. It is difficult to detect unsuitable materials for incineration. On the other hand, since the entire garbage image is searched, it is possible to detect an incineration inappropriate material without omission. For this reason, the example demonstrated by this embodiment is suitable for the execution at night and a holiday which does not carry in garbage.

  In the present embodiment, in order for the unsuitable object detection system 100 to notify the carrying-in main body that an unsuitable incineration object has been detected, based on the detection result, whether the unloadable non-burning material is included in the garbage loaded by the carrying-in main body. Create a report showing whether or not. The associating unit 109 and the printing apparatus 7 that have not been described in the first embodiment execute processing related to the creation of the form.

  In the present embodiment, the inappropriate object detection system 100 is assumed to be a system that executes processes described below in addition to the processes described in the first embodiment. However, the inappropriate object detection system 100 is not limited to this example, and may be a system that executes only one of the processes described in the first embodiment or the processes described in the second embodiment.

  In addition to the functions described in the first embodiment, the garbage image acquisition unit 101 reads out a garbage image captured within a predetermined period from the data server 6 and outputs the garbage image to the image region extraction unit 102.

  In addition to the functions described in the first embodiment, the vehicle image acquisition unit 108 reads a vehicle image captured within a predetermined period from the data server 6 and outputs the vehicle image to the association unit 109. In addition, the garbage image and vehicle image which were image | photographed within the predetermined period may be a garbage image and vehicle image which were image | photographed on the day, when the process demonstrated by this embodiment is performed at night, for example.

  The associating unit 109 sets the garbage carrying subject specified from the vehicle image and the detection result acquired from the inappropriate object detection unit 104 based on the time information and door identification information associated with the vehicle image and the detection result. Associate. Also, the association unit 109 generates a form based on this association. The details of the process of specifying the import subject and the process of associating the import subject with the detection result will be described later.

  The printing device 7 performs printing under the control of the information processing device 1. For example, the printing apparatus 7 prints the above-described form under the control of the association unit 109.

[Process flow]
The flow of processing executed by the information processing apparatus 1 will be described with reference to FIG. 11. FIG. 11 is a flowchart illustrating an example of the flow of processing executed by the information processing apparatus 1.

  The garbage image acquisition unit 101 and the vehicle image acquisition unit 108 read out the garbage image and the vehicle image taken within a predetermined period from the data server 6 (S201). Then, the garbage image acquisition unit 101 outputs the read garbage image to the image region extraction unit 102. The vehicle image acquisition unit 108 outputs the read vehicle image to the association unit 109.

  The image area extraction unit 102 selects one from the read garbage images (S202). For example, the image area extracting unit 102 may select the oldest garbage image that has the oldest date and time indicated by the date and time information. Subsequently, the image area extracting unit 102 searches the entire selected garbage image (S203), and determines whether there is a garbage area (S204). If it is determined that there is a garbage area (YES in S204), the image area extraction unit 102 adds information indicating the garbage area to the garbage image and outputs the garbage image to the garbage identification unit 103. Then, the process shown in FIG. 11 proceeds to step S205. On the other hand, when it is determined that there is no garbage area (NO in S204), the image area extraction unit 102 selects another garbage image (S212). Then, the process shown in FIG. 11 returns to step S203.

(Specific example of the search process for the entire garbage image)
Here, a specific example of the entire dust image search process in step S203 will be described with reference to FIG. FIG. 12 is a diagram illustrating a specific example of search processing for the entire garbage image. In the present embodiment, as an example, a sliding window method is employed to search the entire garbage image. Although the detailed description is omitted because it is an existing technology, the sliding window method is to set a small area of a predetermined area for an image, and shift the small area for each predetermined pixel to change the object candidate area. It is a technique to extract.

  As shown in FIG. 12, the image area extraction unit 102 sets a small area 231 for the selected garbage image 28. Then, as shown in FIG. 12, the image region extraction unit 102 extracts a dust region by moving the small region 231 in the direction of the arrow for each predetermined pixel. The search may be performed by changing the size of the small area 231. Thereby, although the time which processing requires increases further, the high precision of the detection of an incineration inadequate thing is realizable. Further, the entire garbage image search process may be performed by a method other than the sliding window method.

  Again, with reference to FIG. 11, it returns to description of the flow of a process. The garbage identification unit 103 identifies the type of garbage from the garbage area of the acquired garbage image (S205), and outputs the identification result to the inappropriate object detection unit 104. The unsuitable object detection unit 104 determines whether the identified garbage is an unsuitable object for incineration (S206). Note that the processing in steps S205 and S206 is the same as the processing in steps S107 and S108 in FIG. 6 described in the first embodiment, and thus detailed description thereof will not be repeated here. When step S206 ends, the unsuitable object detection unit 104 outputs the determination result (detection result) of step S206 to the association unit 109. If the detection result indicates that an incineration inappropriate material has been detected, the inappropriate material detection unit 104 may output information indicating the type of the detected incineration inappropriate material to the association unit 109 together with the detection result. Good.

  Next, the associating unit 109 identifies a carry-in subject from the vehicle image acquired from the vehicle image acquiring unit 108 (S207). Specifically, the associating unit 109 identifies, from the acquired vehicle image, a vehicle image in which the assigned date and time information and door identification information match the date and time information and door identification information given to the acquired detection result. . Subsequently, the associating unit 109 identifies the number of the garbage truck 200 by performing image recognition on the license plate 900 of the garbage truck 200 that appears in the identified vehicle image. Subsequently, the associating unit 109 refers to the carry-in subject specifying information 112 stored in the storage unit 11. The carry-in subject specifying information 112 is information in which the number of the garbage truck 200 entering and leaving the waste incineration facility is associated with information indicating the carry-in subject. The associating unit 109 refers to the importing subject specifying information 112 and specifies information indicating the importing subject associated with the specified number. Note that the information indicating the import subject is, for example, the name of the import subject.

  Next, the associating unit 109 associates the acquired detection result of the incineration inappropriate object with the specified carrying-in subject (S208), and stores it in the storage unit 11. The association unit 109 may store the association in the association table 113 as an example. The association table 113 is a table that stores the detection result and the import subject associated with each other by the association unit 109, and FIG. 13 is a diagram illustrating a specific example thereof. In the example of FIG. 13, the association unit 109 associates the date and time when the garbage is carried in addition to the detection result and the carrying subject. In the example of FIG. 13, when the detection result indicates that an incineration inappropriate material has been detected, the type of the incineration inappropriate material is further associated with the garbage image in which the incineration inappropriate material is reflected. As date and time to be associated, date and time information given to the acquired detection result may be used. The associating garbage image may be read from the data server 6 by the associating unit 109 based on the date information and the door identification information, or may be output from the unsuitable object detecting unit 104 to the associating unit 109 together with the detection result. .

  In the present embodiment, a plurality of garbage images and vehicle images are generated in one dumping of garbage. That is, a plurality of correspondences are generated for one dumping of garbage. The association unit 109 may store a part of the plurality of associations in the association table 113. Specifically, the associating unit 109 may store an association indicating that the detection result has detected an incineration inappropriate material in the association table 113. Further, the association unit 109 selects one of the plurality of associations when there is no association indicating that the incompatibility of the detection result is detected in the plurality of associations in one dumping of garbage. May be stored in the association table. Thereby, the capacity of data stored in the association table 113 can be suppressed.

  The control unit 10 repeats the processing from step S203 to S209 until the association in step S209 is completed for all the garbage images having the garbage area. Specifically, if the association has not been completed (NO in S209), the image area extraction unit 102 selects another garbage image (S212), and the process shown in FIG. 11 returns to step S203.

  On the other hand, when the association is completed (YES in S209), the association unit 209 generates a form including the presence / absence of an incineration-inappropriate object and the import subject based on the association (S210). In the present embodiment, the association unit 109 refers to each record in the association table 113 shown in FIG. 13, and when the date and time and the detection result indicate that the incineration inappropriate material is detected, the incineration inappropriate material. And a form that further includes a garbage image in which the incineration inappropriate material is shown. The generated form is not limited to this example as long as it includes at least the presence / absence of an incineration-inappropriate substance and the main subject of the import. For example, the form may further include a vehicle image when the detection result indicates that an incineration inappropriate material has been detected. Thereby, it can prove that the carrying-in subject shown in the form dropped the incineration inappropriate material.

  Finally, the associating unit 109 controls the printing device 7 to print the generated form (S211). Thus, the process illustrated in FIG. 11 ends.

[Embodiment 3]
When the unsuitable material detection unit 104 detects an unsuitable material for incineration, it is preferable that the waste containing the unsuitable material for incineration is not thrown into the garbage pit. In the present embodiment, an example of controlling so that garbage containing incineration inappropriate materials is not thrown into the garbage pit will be described with reference to FIG. FIG. 14 is a diagram illustrating a configuration example of a garbage storage unit provided in the front stage of the garbage pit.

  The waste storage section shown in FIG. 14 is a section from a waste carry-in entrance (opening in which the door 300 is provided in FIG. 4 and the like. The door 300 is not shown in FIG. 14) to the waste pit. A door 330 is provided at the opening on the garbage pit side of the garbage storage part. The door 330 is switched between a closed state shown in FIG. 14A and an open state shown in FIG. 14B by external control.

  The bottom of the garbage storage part has an inclined surface, and the garbage thrown in from the carry-in entrance slides down along the inclined surface. When the garbage is thrown in, the door 330 is kept in a closed state, so that the dust sliding down along the inclined surface is blocked by the door 330 and is held in the garbage storage unit (holding state). And by switching the door 330 to an open state, the garbage of a garbage storage part is discharge | released in a garbage pit (discharge state).

  The control unit 10 of the information processing apparatus 1 according to the present embodiment includes a holding control unit that controls opening and closing of the door 330 to switch between the holding state and the releasing state. The holding control unit closes the door 330 until the process of determining whether or not the garbage is an incineration inappropriate material (S108 in FIG. 6) is completed at least after the introduction of the garbage into the garbage storage unit is started. Keep it as In the present embodiment, since the carried-in garbage temporarily stays in the garbage storage unit, an incinerator suitable for incineration may be detected from an image obtained by photographing the garbage held in the garbage storage unit.

  And the holding control part maintains the door 330 in a closed state, when the result of the said determination is a result that garbage is an incineration unsuitable thing. Thereby, since an incineration inappropriate material is hold | maintained in a garbage storage part, when this state is hold | maintained, an incineration inappropriate material can be remove | excluded and it can prevent that an incineration inappropriate material is thrown into a garbage pit.

  On the other hand, if the result of the above determination is that the garbage is not an incineration inappropriate product, that is, the retention control unit has detected that the inappropriate product detection unit 104 has not detected an incineration inappropriate product from the waste holding unit. The door 330 is switched from the closed state to the open state. As a result, the waste holding unit is switched from the holding state to the discharging state, and the waste that has been confirmed to contain no incinerator is put into the waste pit.

  In the present embodiment, the holding state and the discharging state of the dust holding unit are switched by opening and closing the double doors 330, but the configuration for switching is not limited to this example. For example, switching may be performed by opening and closing a single door, or may be switched by a shutter. In addition, for example, the inclination angle of the waste storage unit itself is variably configured so that the waste storage unit has a horizontal or nearly horizontal inclination angle in the holding state, and the waste storage unit has a downward slope toward the garbage pit in the released state. You may make it incline.

[Embodiment 4]
In the present embodiment, an example of detecting an incineration inappropriate material based on sound or vibration will be described based on FIG. FIG. 15 is a block diagram illustrating an example of a configuration of a main part of the information processing apparatus 1 that detects incineration inappropriate materials based on sound and vibration. The information processing apparatus 1 according to the present embodiment is different from the information processing apparatus 1 according to the first embodiment illustrated in FIG. 1 in that the control unit 10 includes a sound detection unit 131 and a vibration detection unit 132. Yes.

  The sound detection unit 131 detects sound generated during a period until the garbage carried into the garbage incineration facility falls into the garbage pit. The sound detected by the sound detector 131 may be, for example, a predetermined pattern. The sound of the predetermined pattern is a sound that is generated when an incineration inappropriate material is thrown in. For example, when an inappropriate metal incinerator is thrown in, when an inadequate incinerator is thrown into the path connecting the outlet and the garbage pit, and on the path (for example, a floor surface (inclined surface)) Metal sound is generated while falling. Also, a metal sound is generated when the incineration inappropriate material falls into the garbage pit. For this reason, the sound detection unit 131 may detect the metal sound as a sound having a predetermined pattern. Metallic sound patterns can be detected by frequency analysis or the like. The generated sound may be acquired by a sound acquisition device such as a microphone and transmitted to the information processing device 1. The sound acquisition device may be installed at an arbitrary position where the above sound can be acquired, such as in the vicinity of the garbage outlet, on the route connecting the outlet and the garbage pit, or near the exit of the route. Note that the sound to be detected by the sound detection unit 131 is not limited to a metal sound as long as it is a sound peculiar to incineration inappropriate materials. Moreover, the sound detection part 131 may distinguish and detect the sound corresponding to each of a plurality of types of incineration inappropriate materials. In this case, a plurality of types of incineration inappropriate materials can be detected. Further, the sound detection unit 131 may detect a specific combination of sounds. For example, a combination of a sound of glass breaking and a sound of glass rubbing against a floor surface or the like may be detected. Thereby, the detection accuracy of glass can be raised.

  The vibration detection unit 132 detects vibration generated during a period until the garbage carried into the garbage incineration facility falls into the garbage pit. The vibration to be detected may be, for example, vibration generated in a path connecting the drop opening and the garbage pit. For example, when the floor surface of the path is an inclined surface, the vibration detection unit 132 may detect vibration generated on the inclined surface when dust slides down the inclined surface. In addition, for example, when a garbage transport device such as a conveyor is provided on the floor surface of the path, the vibration detection unit 132 may detect vibration generated in the transport device. The method for detecting vibration is not particularly limited. For example, a configuration in which a vibration measuring device (vibrometer) is installed on a route connecting a drop opening and a garbage pit and the detected value of the vibration measuring device is transmitted to the information processing device 1 is used. It is good.

  Further, the vibration detected by the vibration detection unit 132 may be a predetermined pattern of vibration. The predetermined pattern of vibration is vibration generated when an incineration inappropriate material is thrown in. For example, the vibration detection unit 132 may detect a high-frequency vibration generated when a metal inadequate incinerator is thrown in as a predetermined pattern of vibration. The high frequency vibration can be detected by high frequency analysis, for example. The unsuitable object detection unit 104 can detect an unsuitable object for incineration based on the detection result. In addition, the vibration which the vibration detection part 133 makes a detection target should just be a vibration peculiar to incineration inappropriate material, and is not restricted to the vibration which a metal generate | occur | produces.

  As described above, the sound detection unit 131 detects the sound generated when an incineration inappropriate material is thrown into the garbage pit, and the vibration detection unit 132 detects the vibration generated when the incineration inappropriate material is thrown into the garbage pit. To do. For this reason, the unsuitable object detection unit 104 of the present embodiment treats the garbage even when the sound detection unit 131 detects a predetermined pattern of sound and when the vibration detection unit 132 detects a predetermined pattern of vibration. Detects that incinerators are included. In other words, the unsuitable object detection unit 104 of the present embodiment is such that at least one of the detection result of the sound detection unit 131, the detection result of the vibration detection unit 132, and the identification result of the dust identification unit 103 includes an unsuitable incinerator. Is determined that there is an incineration inappropriate material. Thereby, it becomes possible to detect an incineration inadequate object that is a detection omission only from a dust image, only sound, or vibration, and the detection accuracy of an incineration inappropriate object can be further increased.

  The unsuitable object detection unit 104 indicates that at least any two of the detection result of the sound detection unit 131, the detection result of the vibration detection unit 132, and the identification result of the dust identification unit 103 indicate that there is an unsuitable incinerator. If it is present, it may be determined that there is an unsuitable for incineration. In addition, the unsuitable object detection unit 104 may determine that an incineration unsuitable object exists when all of these results indicate that an unsuitable incineration object exists.

  Further, the unsuitable object detection unit 104 determines whether or not there is an incineration unsuitable object by combining the detection result of the sound detection unit 131, the detection result of the vibration detection unit 132, and the identification result of the dust identification unit 103. May be. In this case, the unsuitable object detection unit 104 may determine whether or not there is an unsuitable object based on the score for each kind (class) of unsuitable object as in the first embodiment. In this case, the detection results of the sound detection unit 131 and the vibration detection unit 132 are reflected in the score. For example, when the sound detection unit 131 detects a metal sound, the unsuitable object detection unit 104 may add a predetermined numerical value to the score of the class of the unsuitable object that is metal. Thereby, the incineration inappropriate object which is a metallic sound can be detected based on both an image and a sound. Similarly, when the vibration detection unit 132 detects a predetermined pattern of vibrations, the unsuitable object detection unit 104 may add a predetermined numerical value to the score of the class of the unsuitable object that is a metal. The numerical value to be added may be determined in advance.

  In the present embodiment, an example in which incineration inappropriate materials are detected using the detection result of the sound detection unit 131, the detection result of the vibration detection unit 132, and the identification result of the dust identification unit 103 has been described. The result of the above may be used to detect unsuitable for incineration. That is, the information processing apparatus 1 is not suitable for incineration if it includes at least one of the dust identification unit 103 (and the garbage image acquisition unit 101 and the image region extraction unit 102), the sound detection unit 131, and the vibration detection unit 132. An object can be detected. For example, if the information processing apparatus 1 includes the sound detection unit 131 and the unsuitable object detection unit 104, the unsuitable object detection unit 104 includes the unsuitable incineration material in the garbage based on the sound detection result by the sound detection unit 131. It is detected that In addition, if the information processing apparatus 1 includes the vibration detection unit 132 and the unsuitable object detection unit 104, the unsuitable object detection unit 104 includes an unsuitable material for incineration based on the vibration detection result by the vibration detection unit 132. It is detected that

[Modification]
The garbage photographing device 2 may include another camera in addition to the high-speed shutter camera. As an example, the garbage photographing apparatus 2 may include a polarization camera. The information processing apparatus 1 can identify a transparent garbage bag by performing image processing on an image captured by a polarization camera. Thereby, it is possible to detect an incinerator inadequate after eliminating a transparent garbage bag that is not incinerator inadequate. As an example, in the first exemplary embodiment, the garbage identifying unit 103 may perform the above-described image processing to identify and exclude an area where a transparent garbage bag appears in the garbage area. Thereby, waste can be identified in a shorter time. In addition, the image area extraction unit 102 may perform the above-described image processing to identify and exclude an area where a transparent garbage bag is shown from the candidate for the garbage area. As a result, the dust region can be extracted in a shorter time. As a result, it is possible to perform the time for detecting the incineration inappropriate material in a shorter time, and it is possible to realize the detection of the incineration inappropriate material with higher real-time property.

  In addition, when the unsuitable object detection unit 104 detects an unsuitable incinerator, the display control unit 106 controls the display device 5 corresponding to the discharge port where the unsuitable incinerator has been dropped to indicate that the unsuitable incinerator has been detected. It may be displayed.

  In the second embodiment, the associating unit 109 may associate the vehicle image with the detection result instead of or together with the importing subject. In the case of the former example, the associating unit 109 may specify the import subject to be included in the form when the form is generated. In addition, since the method demonstrated in Embodiment 2 should just be used about the identification method of a carrying-in main body, description is not repeated here. In the second embodiment, the associating unit 109 may identify the carry-in subject from the information for identifying the carry-in subject presented when the importer enters the waste incineration facility.

  The information processing apparatus 1 may store only the garbage image and the vehicle image at the time of throwing in the data server 6. Specifically, the garbage image acquisition unit 101 does not have a function of storing a garbage image in the data server 6. Further, the image area extraction unit 102 outputs the garbage image from which the garbage area has been extracted to the vehicle image acquisition unit 108. The vehicle image acquisition unit 108 acquires a vehicle image in which the given date information and door identification information match the date information and door identification information of the garbage image acquired from the image region extraction unit 102 among the acquired vehicle images. The data is stored in the data server 6 together with the waste image. Thereby, the capacity of data stored in the data server 6 can be suppressed.

  The information processing apparatus 1 may be specialized for detection of long objects among incineration inappropriate objects. Specifically, the unsuitable object detection unit 104 may detect, as the unsuitable object for incineration, the garbage having a length equal to or greater than a predetermined value among the garbage that the garbage identifying unit 103 has identified as not being a garbage bag. The determination of whether or not the length of the garbage is greater than or equal to a predetermined value is, for example, whether or not at least one side of the rectangular garbage area extracted by the image area extraction unit 102 has a length greater than or equal to a predetermined value. You can do that. In the case of this example, the garbage identifying unit 103 may identify whether or not the object shown in the garbage area is a garbage bag. That is, in machine learning, it is only necessary to learn only garbage bags, so that the burden of machine learning can be reduced.

  In addition, the information processing apparatus according to one embodiment of the present invention only needs to have at least a function of acquiring a garbage image from the garbage photographing apparatus 2 and a function of detecting an incineration inappropriate object from the garbage image. In other words, the control unit 10 of the information processing apparatus 1 only needs to include at least the garbage image acquisition unit 101 and the inappropriate object detection unit 104. That is, the functions corresponding to the image area extraction unit 102, the dust identification unit 103, the alarm control unit 105, the display control unit 106, the crane control unit 107, the vehicle image acquisition unit 108, and the association unit 109 are the same as the information processing apparatus 1. Another device may have. The information processing apparatus 1 and the other apparatus in this example implement the functions of the information processing apparatus 1 described in the first and second embodiments by appropriately transmitting and receiving information.

[Example of software implementation]
A control block (particularly, each unit included in the control unit 10) of the information processing apparatus 1 may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software. Good.

  In the latter case, the information processing apparatus 1 includes a computer that executes instructions of a program that is software for realizing each function. The computer includes, for example, one or more processors and a computer-readable recording medium storing the program. In the computer, the processor reads the program from the recording medium and executes the program, thereby achieving the object of the present invention. For example, a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) can be used as the processor. As the recording medium, a “non-temporary tangible medium” such as a ROM (Read Only Memory), a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used. Further, a RAM (Random Access Memory) for expanding the program may be further provided. The program may be supplied to the computer via an arbitrary transmission medium (such as a communication network or a broadcast wave) that can transmit the program. Note that one embodiment of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention.

[Additional Notes]
An information processing apparatus according to an aspect of the present invention includes a waste image acquisition unit that acquires a waste image that is taken into a waste pit and is taken in the waste pit, and that is unsuitable for incineration from the waste image. And a detection unit for detecting.

  According to the above configuration, since the incineration inappropriate material is detected from the garbage image, the incineration inappropriate material is detected in comparison with the conventional technique such as detecting the incineration inappropriate material by irradiating the waste collection part of the garbage truck with X-rays. Detection can be realized more easily.

  In the information processing apparatus, a vehicle image acquisition unit that acquires a vehicle image obtained by photographing the vehicle carrying the garbage, the vehicle image, and at least one of the garbage carrying entities specified from the vehicle image, You may further provide the matching part which matches the detection result of a detection part.

  According to said structure, since at least one of a vehicle image and a carrying-in main body is matched with a detection result, the carrying-in main body into which the incineration inappropriate material was thrown in can be specified.

  In the information processing apparatus, the garbage image acquisition unit acquires a plurality of the garbage images in time series until the dropped garbage falls in the garbage pit, and from the acquired garbage images, the garbage images are acquired in time series. An extraction unit that extracts a changing image region may be provided, and the detection unit may detect an incineration inappropriate material from the image region.

  According to the above configuration, since the incineration inappropriate object is detected for the image area changing in time series, the detection target area can be narrowed down. Therefore, the detection speed can be improved.

  The information processing apparatus further includes an identification unit that identifies a type of garbage that appears in the garbage image, and the detection unit detects, as an unsuitable for incineration, garbage whose identification result by the identification unit is a predetermined type. May be.

  According to the above configuration, it is possible to detect a predetermined type of garbage as an incineration inappropriate material.

  In the information processing apparatus, the detection unit may detect the garbage as an incineration inappropriate item when the identification unit has identified that the type is not a garbage bag and has a length greater than or equal to a predetermined value. Good.

  According to said structure, the long garbage which becomes the cause of obstruction | occlusion of ash extraction equipment by remaining unburned can be detected.

  The information processing apparatus includes: a waste holding unit provided in a front stage of the waste pit; a holding state in which the waste carried into a waste incineration facility is held in the waste holding unit; and the waste from the waste holding unit. A holding control unit that switches between a discharge state discharged into the garbage pit, and the holding control unit, when the detection unit does not detect incineration inadequate from the garbage of the garbage holding unit, May be configured to switch from the holding state to the releasing state.

  According to the configuration described above, the waste from which no incinerator is detected is discharged into the garbage pit, so that it is possible to prevent the incinerator from being introduced into the garbage pit. In addition, when an incineration inappropriate material is detected, the incineration inappropriate material stays in the garbage holding unit, and therefore the incineration inappropriate material can be easily collected.

  The information processing apparatus includes a sound detection unit that detects sound generated in a period from when the garbage is carried into the garbage incineration facility until it falls into the garbage pit, and the garbage is carried into the garbage incineration facility. And a vibration detection unit that detects vibration generated in a period until it falls into the garbage pit, and the detection unit includes an analysis result of the garbage image, the sound detection unit, and the sound detection unit. The structure which detects the said incineration inappropriate material based on the detection result of at least any one of a vibration detection part may be sufficient.

  According to said structure, compared with the structure which detects an incineration inappropriate material only based on the analysis result of a garbage image, the detection precision of an incineration inappropriate material can be improved. For example, it is possible to detect an incineration inappropriate object that is not shown in the garbage image from sound or vibration.

  The control device according to an aspect of the present invention can control the operation of a crane that transports garbage in the garbage pit when the information processing apparatus detects the incineration inappropriate material, so that the incineration inappropriate material can be incinerated. Move to a position that can be distinguished from nagomi.

  According to said structure, since an incineration inappropriate material is moved to the position which can be distinguished from the incineratorable waste, it can prevent that an incineration inappropriate material is incinerated.

  The unsuitable object detection system according to one aspect of the present invention is an unsuitable object detection system including an imaging device that captures garbage in the middle of being brought into a garbage incineration facility and dropped into a garbage pit, and an information processing device. The information processing apparatus acquires a garbage image obtained by photographing the garbage from the photographing apparatus, and detects an incineration inappropriate material from the garbage image. According to this unsuitable object detection system, there exists an effect similar to the said information processing apparatus.

  In the unsuitable object detection system, the information processing apparatus further includes an alarm device that performs detection of the unsuitable object for incineration when dropping garbage and outputs an alarm when the information processing apparatus detects the unsuitable object for incineration. But you can.

  According to said structure, it can alert in real time to the person who carried in the incineration inappropriate material. In addition, it is possible to promptly recognize that inappropriate incineration materials have been thrown into the facility workers.

  The unsuitable object detection system may further include a display device that is provided in the vicinity of a dropping port for dropping garbage into the garbage pit and displays the garbage image taken by the photographing device when the garbage is dropped.

  According to the above configuration, it is possible to show a garbage image to the person who brought in the garbage. This makes it possible for the importer to be conscious of being “seen” and to prevent the incineration of unsuitable items.

  Another information processing apparatus according to an aspect of the present invention includes a sound detection unit that detects sound generated during a period from when the garbage carried into the garbage incineration facility falls into the garbage pit after being carried in, and the sound detection unit And an inappropriate material detection unit that detects that the waste contains an inappropriate material for incineration based on the detection result.

  According to said structure, based on the sound generated in the period until the carried-in garbage falls into a garbage pit, it detects that the said garbage contains the incineration inappropriate material. Therefore, detection of incinerators can be more easily realized as compared with conventional techniques such as detecting an incinerator inappropriate object by irradiating the waste collection part of the garbage truck with X-rays.

  Still another information processing apparatus according to an aspect of the present invention includes a vibration detection unit that detects vibration generated in a period from when the garbage carried into the garbage incineration facility falls into the garbage pit after being carried in, and the vibration detection And an inappropriate material detection unit that detects that the waste contains an inappropriate material for incineration based on the detection result of the part.

  According to said structure, based on the vibration generate | occur | produced in the period until the refuse carried in falls into a garbage pit, it detects that the said garbage contains the incineration inappropriate material. Therefore, detection of incinerators can be more easily realized as compared with conventional techniques such as detecting an incinerator inappropriate object by irradiating the waste collection part of the garbage truck with X-rays.

  The information processing apparatus described above may be realized by a computer. In this case, the information processing apparatus that causes the information processing apparatus to be realized by the computer by operating the computer as each unit (software element) included in the information processing apparatus. A program and a computer-readable recording medium on which the program is recorded also fall within the scope of the present invention.

1 Information processing device (control device)
2 Garbage shooting device (shooting device)
4 Alarm Device 5 Display Device 8 Crane 100 Inappropriate Object Detection System 101 Garbage Image Acquisition Unit 102 Image Area Extraction Unit (Extraction Unit)
103 Waste identification part (identification part)
104 Inappropriate object detection part (detection part)
108 vehicle image acquisition unit 109 association unit 131 sound detection unit 132 vibration detection unit

Claims (13)

A waste image acquisition unit for acquiring a waste image taken in the middle of the waste being brought into the waste incineration facility and dropped in the waste pit;
An information processing apparatus, comprising: a detection unit that detects incineration inappropriate materials from the garbage image. A vehicle image acquisition unit for acquiring a vehicle image obtained by photographing the vehicle carrying the garbage;
2. The apparatus according to claim 1, further comprising: an association unit that associates at least one of the vehicle image and the garbage carrying subject specified from the vehicle image with a detection result of the detection unit. Information processing device. The garbage image acquisition unit acquires a plurality of the garbage images in a time series until the dropped garbage falls in the garbage pit,
An extraction unit that extracts an image area that changes in time series from the plurality of acquired garbage images,
The information processing apparatus according to claim 1, wherein the detection unit detects an inappropriate incinerator from the image area. An identification unit for identifying a type of garbage that appears in the garbage image;
The information processing apparatus according to any one of claims 1 to 3, wherein the detection unit detects, as an incineration inappropriate material, garbage whose identification result by the identification unit is a predetermined type.   The said detection part detects the said garbage as an incineration inappropriate object, when the identification part identifies that the said classification is not a garbage bag has a length more than predetermined value. The information processing apparatus described in 1. The waste holding unit provided in the preceding stage of the waste pit is in a holding state in which the waste carried into the waste incineration facility is held in the waste holding unit, and the waste is discharged from the waste holding unit into the waste pit. A holding control unit that switches between the release state and
The said holding | maintenance control part switches the said waste holding part from a holding | maintenance state to a discharge | release state, when the said detection part does not detect incineration inappropriate material from the garbage of the said garbage holding part. The information processing apparatus according to any one of the above. A sound detection unit for detecting a sound generated in a period until the garbage falls into the garbage pit after being carried into the garbage incineration facility;
A vibration detection unit that detects vibration generated in a period until the garbage falls into the garbage pit after being carried into the garbage incineration facility,
The said detection part detects the said incineration inadequate based on the analysis result of the said waste image, and the detection result of at least any one of the said sound detection part and the said vibration detection part. The information processing apparatus according to any one of 6.   When the information processing apparatus according to any one of claims 1 to 7 detects the unsuitable incinerator, the operation of a crane that transports the waste in the waste pit is controlled to incinerate the unsuitable incinerator. A control device that moves to a position that can be distinguished from possible waste. An inadequate object detection system including a photographing device for photographing waste that is brought into a garbage incineration facility and being dropped into a garbage pit, and an information processing device,
The information processing apparatus includes:
Obtaining a garbage image of the garbage from the photographing device,
An unsuitable object detection system for detecting an unsuitable object for incineration from the waste image. The information processing device performs detection of the unsuitable incinerator when dropping garbage,
The improper object detection system according to claim 9, further comprising an alarm device that outputs an alarm when the information processing apparatus detects the incineration inappropriate object.   11. The display device according to claim 9, further comprising a display device that is provided in a vicinity of a dropping port for dropping garbage into the garbage pit and displays the garbage image taken by the photographing device when the garbage is dropped. Inappropriate object detection system. A sound detection unit that detects sound generated during the period from when the garbage brought into the garbage incineration facility falls into the garbage pit after being carried in;
An information processing apparatus comprising: an inappropriate material detection unit that detects that the garbage contains an incineration inappropriate material based on a detection result of the sound detection unit. A vibration detector that detects vibrations that occur during the period from when the garbage carried into the garbage incineration facility falls into the garbage pit after being carried in;
An information processing apparatus comprising: an inappropriate material detection unit configured to detect that the waste contains an incineration inappropriate material based on a detection result of the vibration detection unit.