A Method for Identifying Bare Coal in Remote Sensing Images

Abstract

The present invention relates to the field of remote sensing and geographic information technology, and specifically relates to a method for identifying bare coal in remote sensing images, which includes the following steps: obtaining data and preprocessing the data, obtaining satellite image data of the target area, extracting representative spectral curves of bare coal and land cover types in different target areas, analyzing the color characteristics of bare coal, analyzing the spectral trend characteristics of bare coal, feature merging, merging the color characteristics and spectral trend characteristics of bare coal according to the following formula to obtain the spectral rules of bare coal, using median filtering to denoise the ECR to obtain the final bare coal identification result. Compared with the existing bare coal identification methods, the bare coal identification method proposed according to the bare coal spectral characteristics of different regions in the present invention can not only identify bare coal on a large scale, but also has high identification accuracy.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing and geographic information technology, and particularly to a method for identifying bare coal in remote sensing images. Background Art

[0002] The change of land use in mining areas is a hot topic, so the evaluation of the change of mining area environment has attracted sufficient attention. Bare coal is a unique land use type in coal mining areas, so bare coal can be regarded as an important symbol of coal mining activities. There have been many related studies on the monitoring of land use change in mining areas, but most of them use supervised classification to classify bare coal and other bare lands caused by mining together as industrial and mining land. In addition, some researchers have formulated relevant bare coal spectral rules by analyzing the spectral characteristics of bare coal. However, the formulated spectral rules only have a high recognition accuracy for local areas and lack wide applicability.

[0003] For example, the patent authorization announcement number CN107895136B discloses a method and system for identifying coal mine areas, which is used to identify coal mine areas. The method includes: obtaining remote sensing image data of a target area and measured spectral data of coal in the target area; screening the measured spectral data to obtain spectral data consistent with the bands where the remote sensing image data is located in the measured spectral data as sample spectral data, and the sample spectral data includes training data and test data; using a network training set containing coal and non-coal spectral data, training the training data with a preset extreme learning machine to obtain an optimal ELM feature classification model set with the best classification recognition rate for the training data; using the optimal ELM feature classification model set to classify the remote sensing image data, and obtaining the remote sensing image data classified and recognized as having coal feature data by the optimal ELM feature classification model set in the remote sensing image data as target image data, and the area corresponding to the target image data is the coal mine area.

[0004] As the above patent is the prior art, the formulated spectral rules only have a high recognition accuracy for local areas and lack wide applicability. Summary of the Invention

[0005] In view of this, the present invention provides a method for identifying bare coal in remote sensing images, which can not only identify bare coal on a large scale but also has a high recognition accuracy.

[0006] To solve the above technical problems, the present invention provides a method for identifying bare coal in remote sensing images, including the following steps:

[0007] Obtain data and preprocess the data, obtain satellite image data of the target area, and extract representative spectral curves of bare coal and land cover types in different target areas;

[0008] Analyze the color characteristics of bare coal. The representative formula for the color characteristics of bare coal is:

[0009]

[0010] where , , and represent the reflectance of the blue, green, red, and near-infrared bands respectively, and max represents the maximum value function;

[0011] Analyze the spectral trend characteristics of bare coal. The spectral trend characteristics of bare coal are as follows:

[0012]

[0013] where , , , , , represent the reflectance of the blue, green, red, near-infrared, short-wave infrared 1, and short-wave infrared 2 bands respectively, and max represents the maximum value function;

[0014] Feature combination. Combine the color features and spectral trend characteristics of bare coal according to the following formula to obtain the bare coal index;

[0015]

[0016] where f is a binary function that returns a value of 1 when the corresponding rules are met, otherwise it returns a value of 0. ECR is the bare coal spectral rule, and its value is 0 or 1, where 1 represents bare coal and 0 represents the background land cover type. Based on the ECR of each image, the identification result of bare coal can be obtained;

[0017] Post-processing. In order to remove the salt-and-pepper noise in the identification result, median filtering is used to denoise the ECR to obtain the final bare coal identification result.

[0018] Furthermore, the satellite image data is the Landsat image data to be identified in the study area downloaded from the United States Geological Survey.

[0019] Furthermore, the impact level of the satellite image data is L1T, and it includes a mask file for clouds and shadows. The mask file included in the image data is used to remove clouds and shadows in the image.

[0020] Furthermore, in the step of analyzing the color characteristics of bare coal, the natural color of bare coal is black. Therefore, it appears dark in the true color composite image of the image. At the same time, bare coal also appears dark in the false color composite image of any combination of the blue, green, red, and near-infrared bands. The reason for the above phenomenon is that the reflectance values of bare coal in these 4 bands are relatively low and close.

[0021] Further, the spectral trend feature of raw coal is obtained by summarizing the trend features of the raw coal spectrum based on the change trends of the raw coal spectra in different regions and in combination with a large number of raw coal sample points.

[0022] Further, in the post-processing step, the sliding window size of the median filtering is 3×3.

[0023] The beneficial effects of the above technical solution of the present invention are as follows:

[0024] By analyzing the color features of raw coal and combining with the spectral trend features of raw coal, the present invention combines the color features and trend features of raw coal according to the following formula to obtain the final raw coal spectral rule; in order to remove the salt-and-pepper noise in the recognition result, median filtering is used to denoise the ECR to obtain the final raw coal recognition result. Compared with the existing raw coal recognition methods, the raw coal recognition method proposed according to the spectral characteristics of raw coal in different regions by the present invention can not only recognize raw coal in a large range, but also has high recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a representative spectral curve graph of raw coal and other land cover types in different study areas in the embodiment of the present invention;

[0026] Figure 2 It is a distribution map of the study area in the embodiment of the present invention;

[0027] Figure 3 It is a raw coal recognition result map of the TM / ETM image of the study area in the embodiment of the present invention;

[0028] Figure 4 It is a raw coal recognition result map of the OLI image of the study area in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will combine the accompanying drawings of the embodiments of the present invention Figures 1-4 to clearly and completely describe the technical solutions of the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention fall within the scope of protection of the present invention.

[0030] Embodiment

[0031] As Figures 1-4 shown: A method for identifying raw coal in remote sensing images

[0032] Data Download and Preprocessing: Download Landsat satellite image data to be identified in the study area from the United States Geological Survey (USGS). The image level is L1T and includes mask files such as clouds and shadows. Then, use the mask files included in the data to remove clouds and shadows from the image. Subsequently, extract representative spectral curves of bare coal and other land cover types in different study areas (such as Figure 1 ) for subsequent analysis;

[0033] Analysis of the Color Characteristics of Bare Coal: The natural color of bare coal is black, so it appears dark in the true color composite image of the image. At the same time, bare coal also appears dark in the false color composite images of any combination of the blue, green, red, and near-infrared bands. The reason for the above phenomenon is that the reflectance values of bare coal in these 4 bands are relatively low and similar, as Figure 2 shown. Therefore, based on a large number of analyses of bare coal samples, the following formula is used to represent the color characteristics (CC) of bare coal:

[0034]

[0035] In the formula , , , , , represent the reflectance of the blue, green, red, and near-infrared bands, shortwave infrared 1 band, and shortwave infrared 2 band respectively, and max represents the maximum value function;

[0036] Feature Merging: Merge the color characteristics and trend characteristics of bare coal according to the following formula to obtain the final bare coal index:

[0037]

[0038] In the formula f is a binary function that returns a value of 1 when the corresponding rules are met, otherwise it returns a value of 0. ECR is the spectral rule of bare coal, and its value is 0 or 1, where 1 represents bare coal and 0 represents the background land type. Based on the ECR of each image, the identification result of bare coal can be obtained;

[0039] Post-processing: To remove the salt-and-pepper noise in the identification result, use median filtering (window size 3×3) to denoise the ECR to obtain the final identification result of bare coal.

[0040] Experimental Example

[0041] Such as Figures 2-4 , apply the method of this patent to 8 different study areas worldwide, download a TM / ETM image and an OLI image for each study area, invert the ECR of each image, and extract bare coal using the ECR result.

[0042] In the present invention, unless otherwise clearly specified and defined, for example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components or the interaction relationship between two components. Unless otherwise clearly defined, for those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0043] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for identifying bare coal in remote sensing images, characterized in that, Including the following steps: Obtain data and preprocess the data, obtain satellite image data of the target area, and extract representative spectral curves of bare coal and land cover types in different target areas; Analyze the color characteristics of bare coal. The representative formula for the color characteristics of bare coal is: where ρ Blue , ρ Green , ρ Red and ρ NIR represent the reflectance of the blue, green, red, and near-infrared bands respectively, and max represents the maximum value function; Analyze the trend characteristics of the bare coal spectrum. The trend characteristics of the bare coal spectrum are as follows: where ρ Blue , ρ Green , ρ Red , ρ NIR , ρ SWIR1 , ρ SWIR2 represent the reflectances of the blue, green, red, and near-infrared bands, short-wave infrared 1 band, and short-wave infrared 2 band, respectively, and max represents the maximum value function; Feature combination. Combine the color characteristics and spectral trend characteristics of bare coal according to the following formula to obtain the bare coal index; ECI = f(CC) × f(TC) In the formula, f is a binary function that returns a value of 1 when the corresponding rule is satisfied, otherwise it returns a value of 0. ECI is the bare coal index, and its value is 0 or 1, where 1 represents bare coal and 0 represents the background land type. The recognition result of bare coal can be obtained based on the ECI of each image; Post-processing. In order to remove the salt-and-pepper noise in the recognition result, median filtering is used to denoise the ECI to obtain the final bare coal recognition result.

2. The remote sensing image bare coal recognition method according to claim 1, characterized in that: The satellite image data is Landsat image data to be recognized in the study area downloaded from the United States Geological Survey.

3. The remote sensing image bare coal recognition method according to claim 2, characterized in that: The impact level of the satellite image data is L1T, and it contains a mask file for clouds and shadows. The mask file contained in the image data is used to remove clouds and shadows in the image.

4. The remote sensing image bare coal recognition method according to claim 1, characterized in that: In the step of analyzing the color characteristics of bare coal, the natural color of bare coal is black, so it appears dark in the true color composite image of the image. At the same time, bare coal also appears dark in the false color composite image of any combination of the blue, green, red, and near-infrared bands because the reflectance values of bare coal in these 4 bands are less than the set value and are similar.

5. The method for identifying bare coal in remote sensing images according to claim 1, wherein: The trend characteristics of the bare coal spectrum are summarized based on the change trends of the bare coal spectra in different regions and in combination with a large number of bare coal sample points.

6. The remote sensing image bare coal recognition method according to claim 1, characterized in that: In the post-processing step, the sliding window size of the median filtering is 3×3.

Images