JP4452793B2 - Illegal dumping point detection device, method, and program - Google Patents

Description

  The present invention relates to an illegal dumping point detection device, method, and program for searching for an illegal dumping place of waste.

  The illegal dumping of waste causes serious environmental pollution, and it is important to detect the dumping site early in order to suppress this damage. Conventional methods of detecting illegal dumping places of waste depend on human resources such as notification of neighboring residents and patrols of observers. There is a limit. Moreover, the place where illegal dumping is usually performed is often a place where it is usually difficult to see.

  For these reasons, it is difficult to find illegal dumping places such as waste. For these reasons, a method for exhaustively searching a wide target range is required for effective discovery of illegal dumping locations.

  In addition, an illegal dumping monitoring system that captures and monitors images according to a conventional technique is to monitor a situation by installing a camera on the ground at a point requiring caution or where illegal dumping has already occurred. For this reason, it is not effective for the purpose of finding a new illegal dumping place.

In addition, what is described in patent document 1, patent document 2, and patent document 3 is known as a prior art. The techniques of Patent Literature 1 and Patent Literature 2 are for monitoring illegal dumping of waste at a specific location, and are not effective for finding the above-mentioned new illegal dumping location. Further, the technique described in Patent Document 3 is a technique for assisting the cyclic monitoring of illegal dumping of wastes and the like, and is not effective for the purpose of finding the above-mentioned new illegal dumping point.
JP 2003-132443 A JP 2002-334386 A JP 2002-041633 A

  The present invention has been made in view of the above circumstances, and an object thereof is to provide an illegal dumping point detection device, method, and program capable of finding and detecting a place where waste or the like has been dumped. is there.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and the invention according to claim 1 is directed to a pixel in which the difference between the brightness of adjacent pixels or the average brightness of a region within image data taken from the sky is within a predetermined range. a division processing means for dividing the homogeneous region generated by integrating, on the basis of the processing result of the segmentation processing unit, for the area in the image data, the vegetation index is an index representing the richness of vegetation, Based on the brightness of a predetermined band, the average value is calculated for each pixel, and the vegetation determination means that considers only the region where the average value of the vegetation index is low as a candidate for illegal dumping location, and the processing result of the division processing means Originally, a luminance determination unit that obtains an average luminance value of the region in the image data, and considers only a region where the average luminance value is within a predetermined range as a candidate for illegal dumping location, and a processing result of the division processing unit Based on the calculated variance value regarding the luminance of the plurality of bands of area in the image data, less than 1 or more or variance value exceeds a predetermined first threshold value, the second all the distributed value is a predetermined threshold value Based on the result of processing by the luminance dispersion value determination means and the division processing means for excluding the area of illegal dumping location as candidates , the dispersion value of the vegetation index of each area in the image data is obtained, and the vegetation index Based on the processing result of the vegetation variance value determination means, which considers only the area where the variance value is equal to or less than a predetermined threshold as an illegal dumping location candidate, the area of the area in the image data is obtained, A total area determining unit that considers only a region having an area within a predetermined range as a candidate for illegal dumping location, the vegetation determining unit, the luminance determining unit, the luminance variance value determining unit, and the vegetation variance value determining unit And the total area judgment hand By a dumping position detecting device, characterized in that the display unit areas in the image data that are considered as candidates for illegal dumping places.

  The invention according to claim 2 is the illegal dumping location detection device according to claim 1, wherein before the processing by the division processing means, correction of brightness of the image data, atmospheric correction, geometric correction, The image processing apparatus further includes correction processing means for performing the above.

  The invention according to claim 3 is the illegal dumping location detecting device according to claim 1 or 2, wherein the luminance dispersion value determining means determines the luminance dispersion value to be a visible red band and a visible green band. And a visible blue band and a near-infrared band.

The invention according to claim 4 is an illegal dumping spot detection method used in an illegal dumping spot detection device, wherein the illegal dumping spot detection device uses image data captured from above to obtain the brightness or area of an adjacent pixel. a dividing process in which the difference between the average luminance is divided into regions of homogeneous generated by integrating a pixel within a predetermined, for the area in the image data, the vegetation index is an index representing the richness of vegetation, the average value determined by calculating for each pixel based on the luminance of a predetermined band, only the area average value is low of the vegetation index and candidate consider as vegetation determination process of illegal dumping places, in the area in the image data the average value of the luminance, the average value of the luminance only in the region within a predetermined range and the candidate consider as brightness determination process of illegal dumping places, min brightness of a plurality of bands of area in the image data Calculated values, one or more of the variance value exceeds a predetermined first threshold value, luminance dispersion value determined to exclude the candidate of all the dispersion value dumping places the region less than a predetermined second threshold and process, obtains the dispersion value of the vegetation index of each region in the image data, only a candidate consider as vegetation variance value determination process of the dumping location area is variance of the vegetation index is below a predetermined threshold value, wherein measuring the area of a region in the image data, the area of the candidate consider as the total area determination process of dumping places only a region within a predetermined range, the vegetation determination process, the luminance determination process, the luminance dispersion An illegal dumping point detection comprising: a value determination process; a vegetation dispersion value determination process; and a display process for displaying a region regarded as an illegal dumping point candidate in the total area determination process on a display unit. Is the method.

The invention according to claim 5 generates image data photographed from above in the computer of the illegal dumping point detection apparatus by integrating pixels whose difference from the luminance of adjacent pixels or the average luminance of the region is within a predetermined range. a dividing processing step of dividing the homogeneous regions, for the area in the image data, the average value of the vegetation index is an index representing the richness of vegetation, is calculated for each pixel based on the luminance of a predetermined band A vegetation determination step in which only a region where the average value of the vegetation index is low is regarded as a candidate for illegal dumping location, an average value of luminance of the region in the image data is determined, and the average value of luminance is within a predetermined range A luminance determination step that considers only a region in the image data as a candidate for illegal dumping location, obtains a variance value of the luminance of a plurality of bands in the region in the image data, and one or more variance values are predetermined first A luminance dispersion value step that excludes regions where all variance values are less than a predetermined second threshold value as candidates for illegal dumping locations, and variance values of the vegetation index of each region in the image data A vegetation dispersion value determination step in which only a region where the variance value of the vegetation index is equal to or less than a predetermined threshold is regarded as a candidate for illegal dumping location, an area of the region in the image data is obtained, and the area is in a predetermined range A total area determining step that considers only a region within the area as a candidate for illegal dumping location, the vegetation determining step, the luminance determining step, the luminance variance value determining step, the vegetation variance value determining step, and the total area determining step Is a program for executing a display step of displaying an area regarded as a candidate for an illegal dumping location on the display unit.

  According to the invention of claim 1, claim 4, or claim 5, it is possible to detect illegal dumping places of waste in a wide range based on image data from the sky taken by an artificial satellite or the like. There is. In addition, the know-how that has been accumulated and used by humans based on experience, such as the brightness of the area (brightness and color), the disorder of the area, the shape and area of the area, and the degree of proximity to artifacts, has been ruled. As a result, there is an effect that an illegal dumping location can be found more effectively.

  According to the invention of claim 2, when using image data taken by an artificial satellite or the like, even if the image data is affected by correction by the atmosphere or the artificial satellite itself, There is an effect that it is possible to find a place where illegal dumping may have occurred.

  According to the invention of claim 3, the near infrared band in addition to the visible band is also subject to calculation, and thereby the effect of being able to find a place where illegal dumping may have been performed more reliably. There is.

  First, the basic concept in the present embodiment will be described. In the present embodiment, a computer user detects a dumping location such as industrial waste by causing a computer to analyze image data obtained by imaging an artificial satellite or the like from the sky.

  In this analysis processing, the computer performs processing based on the rules shown in [Table 1]. For example, this rule is that a long and narrow area on the image data is regarded as a road and is determined not to be an illegal dumping place, or an area having a large variation in the vegetation index is determined not to be an illegal dumping place. These rules correspond to know-how when humans visually detect images and detect illegal dumping points.

  FIG. 2 shows the configuration of the computer 100 in this embodiment. The computer 100 stores image data C, which is a ground image taken by an artificial satellite, in a storage area in advance. This image data C is multispectral image data including a visible region, and an R (Red) band that is a visible red band included in reflection from the ground, a G (Green) band that is a visible green band, Information on a B (Blue) band that is a visible blue band and an Ir (Infrared) band that is a near-infrared band is included.

  The control unit 101 is a control function of the computer 100, and details will be described later. The correction processing unit 102 corrects the image data C so that it can be analyzed, and obtains processed image data C. The division processing unit 103 identifies regions of the same quality such as forests and factories in the processed image data C, performs region division, and generates region information that is information about the region. The vegetation determination unit 104 determines the vegetation of each region in the processed image data C, and when there is vegetation, sets the region as the region information that is not subject to subsequent processing. The luminance determination unit 105 sets an area corresponding to a cloud or a water area in the processed image data C as area information to be excluded from subsequent processing. The luminance variance value determination unit 107 determines whether each area is a residential area, farmland, or the like based on the luminance dispersion of each area in the processed image data C, and corresponds to a residential area, farmland, or the like. In this case, the area is set as area information outside the scope of subsequent processing.

  The vegetation dispersion value determination unit 106 determines whether or not each area is an artificial vegetation area based on the vegetation dispersion of each area in the processed image data C. Sets the area in the area information as not subject to subsequent processing. When the area of each region in the processed image data C is extremely large or small, the total area determining unit 108 sets the region information as a target for subsequent processing. Based on the shape of each area in the processed image data C, the shape determination unit 109 determines whether each area is a road, a tunnel, or the like. Set to information. The peripheral area identification unit 110 determines whether or not the peripheral area of each area in the processed image data C is an artifact, and sets the area information as not being an illegal dumping location if applicable.

Next, the flow of processing in the present embodiment will be described with reference to FIG. Now, the user of the computer 100 inputs designation of the image data C to the operation unit of the computer 100 and instructs the computer 100 to display a place where illegal dumping may have occurred.
The control unit 101 of the computer 100 receives a user instruction via the operation unit of the computer 100, and requests the correction processing unit 102 to designate the image data C so that the image data can be analyzed.

  The correction processing unit 102 receives an instruction from the control unit 101 and performs image correction processing corresponding to step S01 in FIG. In this processing, the correction processing unit 102 first reads the image data C from the storage area. Then, the correction processing unit 102 returns this correction because the image data C is sent from an artificial satellite and is corrected so that it can be easily seen by humans. As a result, the image data C becomes an original image captured by the sensor of the artificial satellite.

  Next, the correction processing unit 102 removes the attenuation of luminance and the like due to radiation from the atmosphere existing between the sensor of the artificial satellite that captured the image data C and the object to be imaged. Perform atmospheric correction. Further, the correction processing unit 102 performs geometric correction, which is a correction process for the distortion of the lens of the artificial satellite sensor and the distortion of the image due to the position and orientation of the artificial satellite, on the image data C. After performing such correction, the correction processing unit 102 writes the image data C in the storage area, and outputs the completion of preprocessing to the control unit 101.

  The control unit 101 inputs a notification of completion of preprocessing from the correction processing unit 102, and then instructs the division processing unit 103 to divide the area of the image data C corresponding to the processing in step S02 of FIG. The division processing unit 103 receives an instruction from the control unit 101 and reads the processed image data C from the storage area. Then, for each pixel on the processed image data C, the division processing unit 103 pays attention to a total of four bands, that is, an R band, a G band, a B band, and an Ir band. Divide.

  When the difference between the brightness of the adjacent pixels or the average brightness of the areas falls within the threshold value of each band, the division processing unit 103 integrates them into one area. Create an area. The division processing unit 103 performs this region division for all the pixels on the processed image data C, and generates and stores region information indicating in which region the pixels of the processed image data C are included based on the result. Write to the area, and notify the control unit 101 of the completion of the area division. By this processing, it is possible to distinguish forests, farmland, and the like on the processed image data C as separate areas.

  The control unit 101 receives the notification from the division processing unit 103, and then specifies the processed image data C and the region information to the vegetation determination unit 104 in order to perform a vegetation index average value calculation process corresponding to step S03 in FIG. Request the execution of the vegetation index average value calculation process. Whether the vegetation of each region in the processed image data C is rich or not is determined by the vegetation index average value calculation process.

  In response to a request from the control unit 101, the vegetation determination unit 104 reads the processed image data C and the region information from the storage region. Then, the vegetation determination unit 104 calculates an NDVI (Normalized Difference Vegetation Index) of each pixel for each region in the processed image data C. This NDVI is obtained by a calculation formula of (Ir band-R band) / (Ir band + R band).

  Furthermore, the vegetation determination unit 104 obtains E (NDVI) that is an average value of NDVI for each region. It can be seen that the vegetation is rich in the region where E (NDVI) is high. Then, the vegetation determination unit 104 compares E (NDVI) of each region with a predetermined threshold value T_E_NDVI, and determines that the region where E (NDVI) ≧ T_E_NDVI is “vegetation region”. Conversely, when E (NDVI) <T_E_NDVI, the vegetation determination unit 104 determines that the area is a “non-vegetation area”.

  The vegetation determination unit 104 sets “vegetation area” in the area information for each area determined as “vegetation area”, and writes the area information in the storage area. In the subsequent processing, an area determined as a “vegetation area” is regarded as an area that is not an illegal dumping point. This determination is based on a rule corresponding to the spectrum (spectral radiance) feature in [Table 1]. Then, the vegetation determination unit 104 completes the determination in the control unit 101 when the determination of “vegetation area” or “non-vegetation area” for all the regions of the processed image data C and the storage of the determination result are completed. To be notified.

  The control unit 101 receives the notification from the vegetation determination unit 104, and subsequently performs the luminance average value calculation processing corresponding to step S04 in FIG. 1, so that the processed image data C and the region information are designated and the luminance determination unit 105 receives the luminance. Requests execution of average value calculation processing. This brightness average value calculation process detects areas with extremely high or low brightness among the areas in the processed image data C, and these areas are considered to be clouds, buildings, or water areas and are subject to subsequent processing. Excluded from. This determination is based on rules corresponding to the spectrum (spectral radiance) characteristics in [Table 1].

  In response to an instruction from the control unit 101, the luminance determination unit 105 reads the processed image data C and the region information from the storage region. Then, the luminance determination unit 105 calculates an average value of each of the R band, the G band, the B band, and the Ir band for a region that is not the “vegetation region” in the processed image data C. The luminance determination unit 105 determines that the region where the average value of the luminances of all the bands is higher than a predetermined threshold is a cloud or an artificial object such as a building or a factory. / Artifact "is set as the area information, and the area information is written to the storage area.

  In addition, the luminance determination unit 105 determines that the region where the average value of the luminances of all the bands is lower than a predetermined threshold is the sea, the river, the pond, or the like. Is set in the area information, and the area information is written in the storage area. When the luminance determination unit 105 completes this process for all the regions in the processed image data C that are not “vegetation regions”, the luminance determination unit 105 notifies the control unit 101 of the completion of the processing.

  The control unit 101 receives the notification from the luminance determination unit 105, and subsequently performs the luminance dispersion value calculation processing corresponding to step S05 in FIG. 1, so that the processed image data C and the region information are designated and the luminance dispersion value determination unit 107 is specified. Is requested to execute this process. In this process, the farmland or residential area is determined based on the values of the R band, G band, B band, and Ir band in the region in the processed image data C. This determination is based on a rule corresponding to the texture feature in [Table 1].

  In response to a request from the control unit 101, the luminance dispersion value determination unit 107 reads the processed image data C and the region information from the storage region. Then, the luminance dispersion value determination unit 107 sets the R band, the G band, the B band, and the Ir band for the area in the processed image data C in which no setting such as “water area” is set in the area information. Each variance value is obtained, and when there are one or more variance values exceeding a predetermined first threshold, it is determined that the area is an artificial object such as a residential area, and area information regarding the area is obtained. Is set to “artifact” and the area information is written to the storage area.

  Further, the luminance variance value determination unit 107 determines that the region is flat or farmland for the region where the variance value is less than the predetermined second threshold, sets “flat” in the region information regarding the region, Write information to the storage area. When this process is completed for all areas in the processed image data C, the luminance dispersion value determination unit 107 notifies the control unit 101 of the completion of the process.

  The control unit 101 receives the notification from the luminance dispersion value determination unit 107, and subsequently performs the vegetation index dispersion value calculation process corresponding to step S06 in FIG. 1, so that the processed image data C and the region information are designated and vegetation dispersion is performed. The value determination unit 106 is requested to execute this process. In this process, based on the NDVI variance value in each area in the processed image data C, if this value is larger than a predetermined threshold value, it is determined that the area is an artificial vegetation area. This determination is based on a rule corresponding to the texture feature in [Table 1].

  The vegetation dispersion value determination unit 106 receives the request from the control unit 101 and reads the processed image data C and the region information from the storage region. Then, the vegetation variance value determination unit 106 obtains V (NDVI), which is an NDVI variance value, for each area in the processed image data C in which “vegetation area” or the like is not set in the area information, and this value For a region satisfying V (NDVI) ≧ T_V_NDVI with respect to a predetermined threshold T_V_NDVI, the region is determined to be an artificial vegetation region, and “artificial vegetation region” is set as region information for the region. Write information to the storage area. When this process is completed for all areas in the processed image data C in which “vegetation area” or the like is not set in the area information, the vegetation dispersion value determination unit 106 notifies the control unit 101 of the completion of the process.

  The control unit 101 receives the notification from the vegetation dispersion value determination unit 106, and subsequently performs the total area calculation processing corresponding to step S07 in FIG. 1, so that the processed image data C and the region information are designated and the total area determination unit 108 is requested to execute this process. In this process, an extremely large area or an extremely small area in the processed image data C is excluded from the subsequent processing objects because it is difficult to be an illegal dumping place from a practical viewpoint. This determination is based on a rule corresponding to the region shape feature in [Table 1].

  In response to a request from the control unit 101, the total area determination unit 108 reads the processed image data C and the region information from the storage region. Then, the total area determination unit 108 obtains an area for a region in the processed image data C in which “water area” or the like is not set in the region information. The total area determination unit 108 checks whether the area of the region falls within the range of these thresholds using two thresholds, a predetermined large threshold value and a small threshold value.

  When the area of the area does not fall within the range of these thresholds, the total area determination unit 108 determines that the area cannot be an illegal dumping place such as waste from a practical viewpoint, and the area Is set to “exclude” in the area information, and the area information is written in the storage area. When this process is completed for all regions in the processed image data C, the total area determination unit 108 notifies the control unit 101 of the completion of the process.

  The control unit 101 receives the notification from the total area determining unit 108, and subsequently performs the region shape determination process corresponding to step S08 in FIG. 1, so that the processed image data C and the region information are designated and the shape determining unit 109 is designated. Request execution of this process. In this process, a long and narrow area in the processed image data C is regarded as a road, a tunnel, or the like, and is excluded from subsequent processing targets. This determination is based on a rule corresponding to the region shape feature in [Table 1].

  The shape determination unit 109 receives a request from the control unit 101 and reads the processed image data C and the region information from the storage region. Then, the shape determination unit 109 has a minimum elliptical major axis and a minor axis that can cover the area of the processed image data C in which “water area” or the like is not set in the area information. Find the length with. Further, the shape determination unit 109 calculates the minor axis / major axis to obtain the ratio between the major axis and the minor axis, and when this ratio is smaller than a predetermined threshold, that is, the region has an extremely elongated shape. If there is, it is determined that the area is a road or a dead road, “exclusion” is set in the area information regarding the area, and the area is written in the storage area. Further, the shape determining unit 109 obtains an elliptical area from the obtained major axis and minor axis, and divides the elliptical area by the area area by dividing the ratio of the elliptical area to the focused area area. calculate. This value indicates the degree of grouping of the area, and when the ratio is larger than a predetermined threshold, the shape determination unit 109 determines that the area is a complex linear area such as a cross shape or an annular shape. Then, “exclude” is set in the area information for the area and written to the storage area. When this process is completed for all areas in the processed image data C, the shape determination unit 109 notifies the control unit 101 of the completion of the process.

  The control unit 101 receives the notification from the shape determination unit 109 and subsequently performs the peripheral region identification process corresponding to step S09 in FIG. 1, so that the processed image data C and the region information are designated and the peripheral region identification unit 110 is designated. Request execution of this process. In this process, if the surrounding area is an artifact in the area of the processed image data C, the area is not an illegal dumping place, and is excluded from subsequent processing targets. This determination is based on the rule corresponding to the peripheral region feature in [Table 1].

  The peripheral area identification unit 110 receives the request from the control unit 101 and reads the processed image data C and the area information from the storage area. Then, the peripheral area identification unit 110 is an area in the processed image data C and the area information is not set as “water area” or “artificial object”, and the area around the area is a predetermined ratio or more. If it is surrounded by an area where “artifact” is set, that is, if the surrounding area is surrounded by buildings, etc., and illegal dumping of waste is difficult, the area is illegally dumped Therefore, “exclude” is set in the area information regarding the area, and the area information is written in the storage area. When the peripheral area identifying unit 110 completes this process for all areas in the processed image data C, it notifies the control unit 101 of the completion of the process.

The control unit 101 receives the notification from the peripheral region identification unit 110, and the region in the processed image data C in which “water area”, “artifact”, or the like has not been set in the region information in the processing so far is a waste item. Processed image that is determined to be an illegal dumping location, processed image data C and area information are read from the storage area, and “water area”, “artifact”, etc. are not set in the area information An area in the data C is displayed on the display unit of the computer 100.
The user confirms the display on the computer 100.

  As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to these embodiments, and design changes and the like within a scope not departing from the gist of the present invention are possible. included. For example, in the present embodiment, the luminance dispersion value determination unit 107 performs processing based on a predetermined value, but this may be arbitrarily changed by the user of the computer 100. Since the user can flexibly change the granularity of detection of the illegal dumping location of the computer 100, the detection granularity desired by the user can be obtained. Note that the present invention can also detect improper storage in intermediate processing facilities and final disposal sites.

It is a flowchart of the computer 100 in embodiment of this invention. It is a block diagram of computer 100 in an embodiment of the invention.

Explanation of symbols

100 ... Computer (Illegal dumping point detection device)
DESCRIPTION OF SYMBOLS 101 ... Control part 102 ... Correction processing part 103 ... Division processing part 104 ... Vegetation judgment part 105 ... Luminance judgment part 106 ... Vegetation dispersion value judgment part 107 ... Luminance dispersion value judgment part 108 ... Total area judgment part 109 ... Shape judgment part 110 ... Peripheral area identification part

Claims (5)

A division processing unit that divides image data captured from the sky into homogeneous regions generated by integrating pixels whose difference between the luminance of adjacent pixels or the average luminance of the region is within a predetermined range;
Based on the processing results of the divided processing means, for the area in the image data, the vegetation index is an index representing the richness of vegetation, the average value is calculated for each pixel based on the luminance of a predetermined band Vegetation judging means for obtaining only a region where the average value of the vegetation index is low as a candidate for illegal dumping point,
A luminance determination unit that obtains an average value of the luminance of the region in the image data based on the processing result of the division processing unit, and considers only a region where the average value of the luminance is within a predetermined range as a candidate for illegal dumping location; ,
Based on the processing result of the division processing means, a variance value relating to the luminance of a plurality of bands in the region in the image data is obtained, and one or more variance values exceed a predetermined first threshold or all variance values A luminance variance value determining means for excluding a region where is less than a predetermined second threshold as a candidate for illegal dumping location;
Based on the processing result of the division processing means, the variance value of the vegetation index of each region in the image data is obtained, and only the region where the variance value of the vegetation index is a predetermined threshold value or less is regarded as a candidate for illegal dumping location. Vegetation dispersion value judging means;
Based on the processing result of the division processing unit, the area of the region in the image data is obtained, and a total area determining unit that considers only a region in which the area is within a predetermined range as a candidate for illegal dumping location,
An area in the image data that is regarded as a candidate for illegal dumping by the vegetation determining means, the luminance determining means, the luminance dispersion value determining means, the vegetation dispersion value determining means, and the total area determining means is displayed. Illegal dumping point detection device characterized by displaying on.   2. The illegal dumping point detection according to claim 1, further comprising correction processing means for performing brightness correction, atmospheric correction, and geometric correction of the image data before processing by the division processing means. apparatus. The luminance dispersion value determining means calculates a luminance dispersion value from a visible red band, a visible green band, a visible blue band, and a near-infrared band. The illegal dumping spot detection device described in 1. An illegal dumping spot detection method used in an illegal dumping spot detection device,
The illegal dumping point detection device is
A division process for dividing image data taken from the sky into homogeneous regions generated by integrating pixels whose difference between the luminance of adjacent pixels or the average luminance of the region is within a predetermined range,
For the areas in the image data, the vegetation index is an index representing the richness of vegetation, calculated for each pixel based on the luminance of a predetermined band and the average value calculated, the average low value area of the vegetation index and the candidate and everyone to vegetation decision process of the illegal dumping places only,
The average value of the luminance of the area in the image data, the average value of the brightness and the candidate consider as brightness determination process of dumping places only a region within a predetermined range,
A luminance dispersion value of a plurality of bands in an area in the image data is obtained, and an area in which one or more dispersion values exceed a predetermined first threshold or all dispersion values are less than a predetermined second threshold. Luminance variance determination process that is excluded from candidates for illegal dumping locations,
Obtains the dispersion value of the vegetation index of each region in the image data, only a candidate consider as vegetation variance value determination process of the dumping location area is variance of the vegetation index is below a predetermined threshold value,
Measuring the area of a region in the image data, the area of the candidate consider as the total area determination process of dumping places only a region within a predetermined range,
A display process for displaying on the display unit a region regarded as a candidate for illegal dumping in the vegetation determination process, the luminance determination process, the luminance dispersion value determination process, the vegetation dispersion value determination process, and the total area determination process When,
An illegal dumping spot detection method characterized by comprising : In the computer of the illegal dumping spot detection device,
A division processing step for dividing the image data taken from the sky into homogeneous regions generated by integrating pixels whose difference between the luminance of adjacent pixels or the average luminance of the region is within a predetermined range;
For the areas in the image data, the vegetation index is an index representing the richness of vegetation, calculated for each pixel based on the luminance of a predetermined band and the average value calculated, the average low value area of the vegetation index A vegetation determination step that considers only a candidate for illegal dumping,
A luminance determination step of obtaining an average value of the luminance of the region in the image data, and considering only a region where the average value of the luminance is within a predetermined range as a candidate for illegal dumping,
A luminance dispersion value of a plurality of bands in an area in the image data is obtained, and an area in which one or more dispersion values exceed a predetermined first threshold or all dispersion values are less than a predetermined second threshold. Luminance variance value step excluded from candidates for illegal dumping locations,
A vegetation dispersion value determination step for obtaining a dispersion value of a vegetation index of each region in the image data, and considering only a region where the dispersion value of the vegetation index is equal to or less than a predetermined threshold;
Determining the area of the region in the image data, and determining the total area only in which the area is within a predetermined range is regarded as a candidate for illegal dumping;
A display step of displaying on the display unit a region regarded as a candidate for illegal dumping in the vegetation determination step, the luminance determination step, the luminance dispersion value determination step, the vegetation dispersion value determination step, and the total area determination step. A program to execute and.

Images