Plastic Solid Waste Remote Sensing Automatic Extraction Method and Device

Abstract

The present invention discloses a remote sensing automatic extraction method and device for plastic solid waste, and belongs to the field of satellite remote sensing technology. Based on high-resolution multi-spectral optical image data, the present invention first performs preprocessing, and then combines the ground spectral curve characteristics of different objects to first extract the impermeable surface area, remove vegetation and other objects, and then perform road masking and water body masking to remove roads and water bodies. After that, a plastic solid waste index is constructed to separate plastic waste from impermeable surfaces. The present invention realizes the accurate identification and extraction of plastic solid waste, reduces the false alarm rate and the missed alarm rate; utilizes the advantages of remote sensing technology to realize rapid monitoring of a large area, provides technical support for solid waste environmental supervision, and promotes environmental protection and sustainable development. The automated process of data processing and extraction is realized, which makes up for the shortcomings of manual monitoring, improves work efficiency, and reduces labor costs.

Description

Technical Field

[0001] The present invention relates to the field of satellite remote sensing technology, and particularly to a method and device for automatically extracting plastic solid waste by remote sensing. Background Art

[0002] With the wide use of plastic products, plastic pollution has become a global environmental problem, especially the problem of marine plastic pollution is becoming increasingly serious. Traditional manual monitoring methods are not only inefficient but also difficult to cover vast areas. Remote sensing technology, with its ability to obtain surface information over a large range and quickly, shows great potential in environmental monitoring. However, existing remote sensing technologies still have problems such as insufficient accuracy and susceptibility to interference in the extraction of plastic solid waste, especially in the identification of small, semi-transparent or plastic fragments similar to the environmental background. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a method and device for automatically extracting plastic solid waste by remote sensing, which realizes the large-scale accurate identification and extraction of plastic solid waste, and reduces the false alarm rate and missed alarm rate.

[0004] The present invention provides the following technical solutions:

[0005] A method for automatically extracting plastic solid waste by remote sensing, the method comprising:

[0006] S1: Obtain a remote sensing image covering the area to be studied, the remote sensing image including a blue band, a green band, a red band, and a near-infrared band;

[0007] S2: Preprocess the remote sensing image;

[0008] S3: Extract the impervious surface area according to the preprocessed remote sensing image;

[0009] S4: Exclude the road area and the water area in the impervious surface area to obtain a target area;

[0010] S5: Construct a plastic solid waste index according to the target area;

[0011]

[0012] Wherein, PLI is the plastic solid waste index of any pixel in the target area, and ρ B 、ρ G 、ρ R are the surface reflectivities of the blue band, the green band, and the red band of this pixel respectively;

[0013] S6: Extract plastic solid waste according to the plastic solid waste index of each pixel in the target area and the set plastic solid waste threshold to obtain a plastic solid waste area.

[0014] Further, S3 includes:

[0015] S31: Using the Normalized Difference Vegetation Index (NDVI) to remove vegetated areas from the preprocessed remote sensing image;

[0016] S32: Using the Impervious Surface Index (ISI) to extract impervious surface areas from the remote sensing image after removing vegetated areas.

[0017] Further, S31 includes:

[0018] S311: Calculating the Normalized Difference Vegetation Index (NDVI) for each pixel of the preprocessed remote sensing image;

[0019]

[0020] ρ NIR is the surface reflectance of the near-infrared band of the pixel of the remote sensing image;

[0021] S312: Comparing the Normalized Difference Vegetation Index of each pixel with the set vegetation index threshold, and classifying the pixels with a Normalized Difference Vegetation Index greater than the vegetation index threshold as vegetated areas;

[0022] S313: Removing the vegetated areas from the preprocessed remote sensing image.

[0023] Further, S32 includes:

[0024] S321: Calculating the Impervious Surface Index (ISI) for each pixel of the remote sensing image after removing vegetated areas;

[0025]

[0026] S322: Comparing the Impervious Surface Index of each pixel with the set impervious surface threshold range, and classifying the pixels with an Impervious Surface Index within the impervious surface threshold range as impervious surface areas.

[0027] Further, S4 includes:

[0028] S41: Masking the roads in the impervious surface area based on road network data to remove the road areas;

[0029] S42: Masking the water bodies in the area after removing the roads through the Normalized Difference Water Index (NDWI) to remove the water body areas and obtain the target area.

[0030] Further, the remote sensing image is a high-resolution optical remote sensing image, and S2 includes:

[0031] Performing radiometric calibration, atmospheric correction, orthorectification, and image enhancement on the remote sensing image.

[0032] Further, S6 includes:

[0033] Compare the plastic solid waste index of each pixel with the plastic solid waste threshold, and classify the pixels with a plastic solid waste index greater than the plastic solid waste threshold as plastic solid waste areas.

[0034] An automatic remote sensing extraction device for plastic solid waste, the device includes:

[0035] A data preparation module for obtaining a remote sensing image covering the area to be studied, the remote sensing image including a blue band, a green band, a red band, and a near-infrared band;

[0036] A preprocessing module for preprocessing the remote sensing image;

[0037] An impervious surface extraction module for extracting an impervious surface area from the preprocessed remote sensing image;

[0038] A masking module for removing the road area and the water area in the impervious surface area to obtain a target area;

[0039] A plastic solid waste index calculation module for constructing a plastic solid waste index based on the target area;

[0040]

[0041] where PLI is the plastic solid waste index of any pixel in the target area, and ρ B , ρ G , ρ R are the surface reflectances of the blue band, the green band, and the red band of this pixel respectively;

[0042] A plastic solid waste extraction module for extracting plastic solid waste based on the plastic solid waste index of each pixel in the target area and the set plastic solid waste threshold to obtain a plastic solid waste area.

[0043] Further, the impervious surface extraction module includes:

[0044] A vegetation removal unit for removing the vegetation area from the preprocessed remote sensing image using the normalized difference vegetation index;

[0045] An impervious surface area extraction unit for extracting an impervious surface area from the remote sensing image after removing the vegetation area using the impervious surface index.

[0046] Further, the vegetation removal unit includes:

[0047] A normalized difference vegetation index calculation sub-unit for calculating the normalized difference vegetation index NDVI of each pixel in the preprocessed remote sensing image;

[0048]

[0049] ρ NIR is the surface reflectance of the near-infrared band of the pixel of the remote sensing image;

[0050] The vegetation area classification subunit is used to compare the normalized vegetation index of each pixel with the set vegetation index threshold respectively, and classify the pixels with a normalized vegetation index greater than the vegetation index threshold as the vegetation area;

[0051] The vegetation area removal unit is used to remove the vegetation area from the preprocessed remote sensing image.

[0052] Furthermore, the impervious surface area extraction unit includes:

[0053] The impervious surface index calculation subunit is used to calculate the impervious surface index VNISI of each pixel of the remote sensing image after removing the vegetation area;

[0054]

[0055] The impervious surface area extraction subunit is used to compare the impervious surface index of each pixel with the set impervious surface threshold range respectively, and classify the pixels with an impervious surface index belonging to the impervious surface threshold range as the impervious surface area.

[0056] Furthermore, the mask module includes:

[0057] The road mask unit is used to mask the roads in the impervious surface area based on the road network data and remove the road area;

[0058] The water body mask unit is used to mask the water body of the area after removing the road through the normalized water body index and remove the water body area to obtain the target area.

[0059] Furthermore, the remote sensing image is a high-resolution optical remote sensing image, and the preprocessing module includes:

[0060] Performing radiometric calibration, atmospheric correction, orthorectification, and image enhancement on the remote sensing image.

[0061] Furthermore, the plastic solid waste extraction module includes:

[0062] Comparing the plastic solid waste index of each pixel with the plastic solid waste threshold, and classifying the pixels with a plastic solid waste index greater than the plastic solid waste threshold as the plastic solid waste area.

[0063] The present invention has the following beneficial effects:

[0064] The present invention realizes the accurate identification and extraction of plastic solid waste, reducing the false alarm rate and missed detection rate; taking advantage of remote sensing technology, it realizes the rapid monitoring of a large area, provides technical support for the environmental supervision of solid waste, and promotes environmental protection and sustainable development. It realizes the automated process of data processing and extraction, makes up for the deficiencies of manual monitoring, improves work efficiency, and reduces labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a flowchart of the method for automatically extracting plastic solid waste by remote sensing according to the present invention;

[0066] Figure 2 is a schematic diagram of the obtained remote sensing image;

[0067] Figure 3 is Figure 2 a schematic diagram of the impervious surface area calculated from the remote sensing image of;

[0068] Figure 4 is Figure 2 a schematic diagram of the plastic solid waste area extracted from the remote sensing image of;

[0069] Figure 5 is a schematic diagram of the surface reflectance curve of typical ground objects;

[0070] Figure 6 is a schematic diagram of the device for automatically extracting plastic solid waste by remote sensing according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0072] Embodiment 1:

[0073] The embodiment of the present invention provides a method for automatically extracting plastic solid waste by remote sensing. As Figures 1-5 shown, the method includes:

[0074] S1: Obtain a remote sensing image covering the area to be studied. The remote sensing image is a high-resolution optical remote sensing image, such as common high-resolution optical images like GF-1, GF-2, GF-6, etc., which includes a blue band, a green band, a red band, and a near-infrared band.

[0075] S2: Preprocess the remote sensing image.

[0076] The preprocessing may include radiometric calibration, atmospheric correction, orthorectification, and image enhancement, etc.

[0077] S3: Extract the impervious surface area according to the preprocessed remote sensing image.

[0078] An impervious surface refers to a surface that does not allow water to penetrate, usually referring to hard surfaces such as buildings, roads, squares, parking lots, etc., as well as land covered with artificial materials and urban construction areas. The plastic solid waste area belongs to a type of impervious surface, so it is necessary to extract the impervious surface area.

[0079] The present invention does not limit the extraction method of the impervious surface area. In one example, it includes the following steps:

[0080] S31: Use the Normalized Difference Vegetation Index to remove the vegetation area from the preprocessed remote sensing image.

[0081] The present invention uses the Normalized Difference Vegetation Index (NDVI) to remove the vegetation-covered area from the study area, thereby extracting the land not covered by vegetation.

[0082] The specific vegetation extraction steps include:

[0083] S311: Calculate the Normalized Difference Vegetation Index NDVI of each pixel of the preprocessed remote sensing image.

[0084]

[0085] ρ NIR and ρ R are the surface reflectance of the near-infrared band and the red band of the pixel of the remote sensing image.

[0086] S312: Compare the Normalized Difference Vegetation Index of each pixel with the set vegetation index threshold respectively, and classify the pixels with a Normalized Difference Vegetation Index greater than the vegetation index threshold as the vegetation area.

[0087] S313: Remove the vegetation area from the preprocessed remote sensing image.

[0088] S32: Use the impervious surface index to extract the impervious surface area from the remote sensing image after removing the vegetation area.

[0089] The present invention extracts the impervious surface area based on a specially constructed impervious surface index. The specific method is as follows:

[0090] S321: Calculate the impervious surface index VNISI of each pixel of the remote sensing image after removing the vegetation area.

[0091]

[0092] S322: Compare the impervious surface index of each pixel with the set impervious surface threshold range respectively, and classify the pixels with an impervious surface index belonging to the impervious surface threshold range as the impervious surface area.

[0093] The remote sensing image before extracting the impervious surface area is as Figure 2 shown, Figure 2 The area enclosed by the red border in Figure 2 is the known plastic solid waste area. According to Figure 3 shown, Figure 3 The part outside the white area in

[0094] S4: Remove the road area and water area in the impervious surface area to obtain the target area.

[0095] Specifically, it includes the following steps:

[0096] S41: Mask the roads in the impervious surface area based on the road network data to remove the road area.

[0097] The road network data can be obtained from the geographic information system. After obtaining the road network data, import the road network data and mask the roads in the image.

[0098] S42: Perform water body masking on the area after removing the roads through the normalized difference water index to remove the water area and obtain the target area.

[0099] Since the water body will be confused with the building shadows in the city during classification, water body masking is performed to remove the water body. In the present invention, the water area is removed by the normalized difference water index and the normalized difference vegetation index. The specific implementation method is as follows:

[0100] S421: Calculate the normalized difference water index NDWI of each pixel of the remote sensing image.

[0101]

[0102] S422: Compare the normalized difference water index of each pixel with the set water index threshold respectively, and classify the pixels with the normalized difference water index greater than the water index threshold as the water area.

[0103] S5: Construct a plastic solid waste index according to the target area.

[0104] The spectral reflection characteristics of each ground object in the extracted target area are as Figure 5 shown. It can be seen that the slopes of the green - red bands of the bare land and the red - roofed buildings are in the opposite state to the slopes of other ground objects. Therefore, the red - roofed buildings and the bare land can be removed by the slopes between different bands. Further analysis shows that the slope of the blue - red band of the plastic solid waste is less than that of the blue - roofed buildings and greater than that of the roads. Therefore, the target ground objects can be extracted by the difference in the blue - red band slopes. That is:

[0105]

[0106] Among them, ρ Rbarren and ρ Gbarren are the reflectance of bare land in the red band and the green band respectively, ρ Bbuild and ρ Rbuild are the reflectance of buildings in the blue band and the red band respectively, ρ Broad and ρ Rroad are the reflectance of roads in the blue band and the red band respectively, ρ Bplastic and ρ Gplastic and ρ Rplastic are the reflectance of plastic solid waste in the blue band, the green band and the red band respectively.

[0107] Thus, a plastic leakage extraction index is constructed, .

[0108] Among them, PLI is the plastic solid waste index of any pixel in the target area, and ρ B , ρ G , and ρ R are the surface reflectance of the blue band, the green band and the red band of this pixel respectively. Multiply by 10 4 constant after the formula to expand the discrimination degree of different ground objects after calculation.

[0109] S6: Extract plastic solid waste according to the plastic solid waste index of each pixel in the target area and the set plastic solid waste threshold to obtain the plastic solid waste area.

[0110] Specifically: Compare the plastic solid waste index of each pixel with the plastic solid waste threshold, and classify the pixels with a plastic solid waste index greater than the plastic solid waste threshold as the plastic solid waste area. Exemplarily, the plastic solid waste threshold can be taken as 16000. According to Figure 2 The extracted plastic solid waste area is as Figure 4 shown, Figure 4 the blue area in

[0111] Based on high-resolution multi-spectral optical image data, the present invention first performs preprocessing, and then combines the ground spectral curve characteristics of different ground objects to first extract the impervious surface area, eliminate ground objects such as vegetation, and then perform road masking and water body masking to remove roads and water bodies. Then a plastic solid waste index is constructed to separate plastic garbage from the impervious surface.

[0112] The present invention realizes the accurate identification and extraction of plastic solid waste, reduces the false alarm rate and the missed alarm rate; utilizes the advantages of remote sensing technology to realize the rapid monitoring of large-scale areas, provides technical support for solid waste environmental supervision, and promotes environmental protection and sustainable development. It realizes the automated process of data processing and extraction, makes up for the deficiencies of manual monitoring, improves work efficiency, and reduces labor costs.

[0113] Example 2:

[0114] An embodiment of the present invention provides a device for automatically extracting plastic solid waste by remote sensing, as Figure 6 shown. The device includes:

[0115] A data preparation module 1 for obtaining remote sensing images covering the area to be studied, where the remote sensing images include a blue band, a green band, a red band, and a near-infrared band.

[0116] A preprocessing module 2 for preprocessing the remote sensing images.

[0117] An impervious surface extraction module 3 for extracting impervious surface areas based on the preprocessed remote sensing images.

[0118] A masking module 4 for removing road areas and water body areas in the impervious surface areas to obtain a target area.

[0119] A plastic solid waste index calculation module 5 for constructing a plastic solid waste index based on the target area.

[0120]

[0121] where PLI is the plastic solid waste index of any pixel in the target area, and ρ B , ρ G , ρ R are the surface reflectances of the blue band, the green band, and the red band of the pixel, respectively.

[0122] A plastic solid waste extraction module 6 for extracting plastic solid waste based on the plastic solid waste indices of the respective pixels in the target area and a set plastic solid waste threshold to obtain a plastic solid waste area.

[0123] The present invention realizes the accurate identification and extraction of plastic solid waste, reduces the false alarm rate and the missed alarm rate; takes advantage of remote sensing technology to achieve rapid monitoring of large-scale areas, provides technical support for solid waste environmental supervision, and promotes environmental protection and sustainable development. It realizes an automated process for data processing and extraction, makes up for the deficiencies of manual monitoring, improves work efficiency, and reduces labor costs.

[0124] As an improvement of the embodiment of the present invention, the aforementioned impervious surface extraction module includes:

[0125] A vegetation removal unit for removing vegetation areas from the preprocessed remote sensing images using the normalized difference vegetation index.

[0126] An impervious surface area extraction unit for extracting impervious surface areas from the remote sensing images after removing the vegetation areas using the impervious surface index.

[0127] Specifically, the vegetation removal unit includes:

[0128] A normalized difference vegetation index (NDVI) calculation subunit, which is used to calculate the NDVI of each pixel of the preprocessed remote sensing image.

[0129]

[0130] ρ NIR is the surface reflectance of the near-infrared band of the pixel of the remote sensing image.

[0131] A vegetation area classification subunit, which is used to compare the NDVI of each pixel with the set vegetation index threshold respectively, and classify the pixels with NDVI greater than the vegetation index threshold as vegetation areas.

[0132] A vegetation area removal unit, which is used to remove the vegetation areas from the preprocessed remote sensing image.

[0133] Correspondingly, the impervious surface area extraction unit includes:

[0134] An impervious surface index calculation subunit, which is used to calculate the impervious surface index (VNISI) of each pixel of the remote sensing image after removing the vegetation areas.

[0135]

[0136] An impervious surface area extraction subunit, which is used to compare the impervious surface index of each pixel with the set impervious surface threshold range respectively, and classify the pixels with impervious surface index belonging to the impervious surface threshold range as impervious surface areas.

[0137] As another improvement of the embodiment of the present invention, the mask module includes:

[0138] A road mask unit, which is used to mask the roads in the impervious surface area based on road network data and remove the road areas.

[0139] A water body mask unit, which is used to mask the water body in the area after removing the roads through the normalized difference water index and remove the water body areas to obtain the target area.

[0140] The remote sensing image in the present invention is a high-resolution optical remote sensing image. Correspondingly, the preprocessing module includes:

[0141] Performing radiometric calibration, atmospheric correction, orthorectification and image enhancement on the remote sensing image.

[0142] As an example, the plastic solid waste extraction module includes:

[0143] Compare the plastic solid waste index of each pixel with the plastic solid waste threshold, and classify the pixels with a plastic solid waste index greater than the plastic solid waste threshold as plastic solid waste areas.

[0144] The device provided by the embodiment of the present invention has the same implementation principle and the same technical effects as those of the foregoing method embodiment. For the sake of brief description, for the parts not mentioned in the device embodiment, reference may be made to the corresponding content in the foregoing method embodiment 1. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the foregoing described device and unit can all refer to the corresponding processes in the foregoing method embodiment, and will not be repeated here.

[0145] Finally, it should be noted that the above-mentioned embodiments are only specific implementation manners of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify or easily conceive of changes to the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention.

Claims

1. A method for automatically extracting plastic solid waste by remote sensing, characterized in that, The method includes: S1: Obtain a remote sensing image covering the area to be studied, where the remote sensing image includes a blue band, a green band, a red band, and a near-infrared band; S2: Preprocess the remote sensing image; S3: Extract the impervious surface area based on the preprocessed remote sensing image; S4: Exclude the road area and water area in the impervious surface area to obtain the target area; S5: Construct a plastic solid waste index based on the target area; where PLI is the plastic solid waste index of any pixel in the target area, and ρ B , ρ G , ρ R are the surface reflectance of the blue band, green band, and red band of this pixel, respectively; S6: Extract plastic solid waste according to the plastic solid waste index of each pixel in the target area and the set plastic solid waste threshold to obtain the plastic solid waste area.

2. The method for automatically extracting plastic solid waste by remote sensing according to claim 1, characterized in that The S3 includes: S31: Use the normalized difference vegetation index to exclude the vegetation area from the preprocessed remote sensing image; S32: Extract the impervious surface area from the remote sensing image after excluding the vegetation area using the visible and near-infrared subpixel impervious surface index.

3. The method for automatically extracting plastic solid waste by remote sensing according to claim 2, wherein The S31 includes: S311: Calculate the normalized difference vegetation index NDVI of each pixel in the preprocessed remote sensing image; ρ NIR is the surface reflectance of the near-infrared band of the pixel of the remote sensing image; S312: Compare the normalized difference vegetation index of each pixel with the set vegetation index threshold, and classify the pixels with a normalized difference vegetation index greater than the vegetation index threshold as the vegetation area; S313: Exclude the vegetation area from the preprocessed remote sensing image.

4. The method for automatically extracting plastic solid waste by remote sensing according to claim 3, wherein, The S32 includes: S321: Calculate the visible and near-infrared subpixel impervious surface index VNISI of each pixel in the remote sensing image after excluding the vegetation area; S322: Compare the visible and near-infrared subpixel impervious surface index of each pixel with the set impervious surface threshold range, and classify the pixels with a visible and near-infrared subpixel impervious surface index belonging to the impervious surface threshold range as the impervious surface area.

5. The method for automatically extracting plastic solid waste by remote sensing according to any one of claims 1-4, characterized in that, The S4 includes: S41: Mask the roads in the impervious surface area based on road network data to remove the road area; S42: Perform water body masking on the area after removing the roads through the normalized difference water index to remove the water area and obtain the target area.

6. The method for automatically extracting plastic solid waste by remote sensing according to claim 5, wherein The remote sensing image is a high-resolution optical remote sensing image, and the S2 includes: Perform radiometric calibration, atmospheric correction, orthorectification, and image enhancement on the remote sensing image.

7. The method for automatically extracting plastic solid waste by remote sensing according to claim 6, wherein The S6 includes: Compare the plastic solid waste index of each pixel with the plastic solid waste threshold, and classify the pixels with a plastic solid waste index greater than the plastic solid waste threshold as the plastic solid waste area.

8. An automatic remote sensing extraction device for plastic solid waste, characterized in that, The device includes: A data preparation module for obtaining a remote sensing image covering the area to be studied, where the remote sensing image includes a blue band, a green band, a red band, and a near-infrared band; A preprocessing module for preprocessing the remote sensing image; An impervious surface extraction module for extracting the impervious surface area based on the preprocessed remote sensing image; A masking module for excluding the road area and water area in the impervious surface area to obtain the target area; A plastic solid waste index calculation module for constructing a plastic solid waste index based on the target area; Among them, PLI is the plastic solid waste index of any pixel in the target area, and ρ B , ρ G , ρ R are the surface reflectance of the blue band, green band and red band of this pixel respectively; A plastic solid waste extraction module for extracting plastic solid waste according to the plastic solid waste index of each pixel in the target area and the set plastic solid waste threshold to obtain the plastic solid waste area.

9. The plastic solid waste remote sensing automatic extraction device according to claim 8, characterized in that The impervious surface extraction module includes: A vegetation exclusion unit for using the normalized difference vegetation index to exclude the vegetation area from the preprocessed remote sensing image; An impervious surface area extraction unit, which is used to extract the impervious surface area from the remotely sensed image after removing the vegetation area by using the impervious surface index.

10. The plastic solid waste remote sensing automatic extraction device according to claim 8 or 9, characterized in that The mask module includes: A road mask unit, which is used to mask the roads in the impervious surface area based on the road network data to remove the road areas; A water body mask unit, which is used to mask the water body in the area after removing the roads by using the normalized difference water index to remove the water body areas and obtain the target area.

Images