A method for evaluating the currentness of vector map data based on multi-temporal remote sensing images

Abstract

The present invention provides a method for assessing the currentness of vector map data based on multi-time series remote sensing imagery, relating to the field of geographic information data processing technology. The method comprises the following steps: acquiring multiple digital orthophoto image data in a time series to generate multi-time series vector data; performing a comparative analysis of settlement elements between the vector map data to be assessed and the vector data from the multi-time series to obtain similarities in the number of settlement elements, the total area of ​​settlement elements, and the geometric similarities of settlement elements between the vector map data to be assessed and the vector data from each time series; calculating a weighted average score between the vector map data to be assessed and the vector data from each time series based on the similarities in the number of settlement elements, the total area of ​​settlement elements, and the geometric similarities of settlement elements; and analyzing the weighted average score to obtain a final assessment result of currentness. This solution improves the accuracy and automation of currentness assessment by utilizing remote sensing image feature extraction technology.

Description

Technical Field

[0001] The present invention relates to the field of geographic information data processing technology, and in particular to a method for evaluating the currentness of vector map data based on multi-temporal remote sensing images. Background Art

[0002] With the rapid development of Earth observation technology, mobile internet technology, satellite positioning technology, and remote sensing image processing technology, the technical difficulty and threshold for obtaining vector map data are becoming increasingly lower. This has led to the emergence of multiple versions of vector map data for the same area. Determining the currency of data is a crucial step in the production and updating of vector map data. The purpose of currency assessment is to determine the currency of vector map data, provide a scientific basis for selecting appropriate data, and incorporate the most current vector map data into vector map data production and updating, thereby improving the quality of the resulting data.

[0003] The timeliness assessment of vector map data is a data analysis process. Common existing methods for assessing the timeliness of vector map data include timestamp analysis, field verification, and manual multi-source data comparison.

[0004] Timestamp analysis: This method analyzes timestamp fields (such as creation time and modification time) in vector map data files and the update interval of statistical data to determine the latest status of the data. Its advantages are simplicity and directness, allowing for quick identification of data update times. Its disadvantages are its over-reliance on timestamps, making it inappropriate for assessing the currency of vector map data that does not contain timestamps.

[0005] Field verification: This involves conducting field surveys of representative features and comparing the results with the accuracy of vector map data to assess the data's currency. This approach offers advantages in accuracy and reliability, allowing for direct verification of data currency. However, it is time-consuming and labor-intensive, making it difficult to adapt to large-scale applications.

[0006] Manual multi-source data comparison: By collecting data from multiple sources, ensuring that the time range of the data source covers the time period of the target data, manual comparison and analysis of vector map feature data and target ground feature data are performed to determine the consistency of the ground feature elements and determine the timeliness of the data based on experience. The advantage is that it can integrate information from multiple sources and improve the accuracy of the timeliness assessment. The disadvantage is that manual analysis is time-consuming and labor-intensive, requires the construction of a large amount of reference data, and the accuracy of the results is significantly affected by the operator's personal experience, making it difficult to handle large-scale data assessments.

[0007] Through the above analysis, the existing vector map data timeliness assessment method relies on manual interaction and operator experience. The process is complex, the degree of automation is low, and the efficiency is low. It is difficult to meet the needs of vector map data production and updating. An automated vector map data timeliness assessment method is urgently needed. Summary of the Invention

[0008] In view of this, an embodiment of the present invention provides a method for assessing the timeliness of vector map data based on multi-temporal remote sensing images to solve the technical problem of difficulty in assessing the timeliness of large quantities of vector map data in the prior art. The method includes:

[0009] Acquire multiple digital orthophoto data in a time series, extract residential features from the multiple digital orthophoto data in the time series respectively, and generate multi-time series vector data, wherein the multi-time series vector data is consistent with the spatial range of the vector map data to be evaluated, and the multi-time series vector data includes multiple single-time series vector data;

[0010] Comparative analysis of settlement elements between the vector map data to be evaluated and the vector data of multiple time series is performed to obtain the similarity in the number of settlement elements, the similarity in the total area of ​​settlement elements, and the geometric similarity of settlement elements between the vector map data to be evaluated and the vector data of each time series;

[0011] Set the similarity weight of the number of settlement elements, the similarity weight of the total area of ​​settlement elements, and the geometric similarity weight of settlement elements. Calculate the weighted average score of the vector map data to be evaluated and the vector data of each time series based on the similarity weight of the number of settlement elements, the similarity weight of the number of settlement elements, the similarity weight of the total area of ​​settlement elements, the similarity weight of the total area of ​​settlement elements, the geometric similarity weight of settlement elements, and the geometric similarity weight of settlement elements.

[0012] The final evaluation result of the timeliness is obtained by analyzing the weighted average score results of the vector data of multiple time series.

[0013] Compared with the prior art, the at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects:

[0014] The vector map data timeliness assessment method of the embodiment of the present invention uses remote sensing image feature extraction technology to automatically assess the timeliness of vector map data. Compared with traditional vector map data timeliness assessment methods, it improves the accuracy and automation of timeliness assessment, and solves the problems of low assessment efficiency, low automation, unquantifiable assessment results, and difficulty in coping with large-scale applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 This is a flow chart of a method for assessing the currentness of vector map data based on multi-temporal remote sensing images provided by an embodiment of the present invention;

[0017] Figure 2 This is a flow chart of a method for implementing the above-mentioned vector map data currentness assessment method based on multi-temporal remote sensing images provided by an embodiment of the present invention;

[0018] Figure 3 It is a thumbnail of vector data of a time series provided by an embodiment of the present invention;

[0019] Figure 4 It is a schematic diagram of the superposition of the vector data to be evaluated and the digital orthophoto data of a known time series. DETAILED DESCRIPTION

[0020] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0021] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the features in the following embodiments and embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of this application.

[0022] The assessment of vector map data currency is a crucial step in map data quality assessment. It not only impacts the validity and reliability of the data but also directly impacts its practicality and application effectiveness. Effective data currency assessment ensures that vector map data provides accurate information across various applications and analyses, supporting better decision-making and spatial analysis.

[0023] An embodiment of the present invention provides a method for assessing the currentness of vector map data based on multi-time series remote sensing images. An artificial intelligence method is used to extract settlement elements from multi-time series digital orthophoto data of known time to obtain multi-time series vector data. According to the shooting time of the known digital orthophoto data, the corresponding generated vector data is sorted and assigned values ​​according to the time axis to obtain vector data containing time information. A multi-time series change analysis is performed, including similarity analysis of the number of settlement elements, similarity analysis of the total area of ​​settlement elements, and geometric similarity analysis of settlement elements. A comprehensive similarity analysis is performed based on the quantitative analysis results to finally determine to which time series the vector map data to be evaluated belongs.

[0024] In an embodiment of the present invention, a method for evaluating the current status of vector map data based on multi-time series remote sensing images is provided. Figure 1 and Figure 2 As shown, the method includes:

[0025] Step S101: Acquire multiple digital orthophoto data in a time series, extract residential area features from the multiple digital orthophoto data in the time series, and generate multi-time series vector data, wherein the multi-time series vector data is consistent with the spatial range of the vector map data to be evaluated, and the multi-time series vector data includes multiple single-time series vector data;

[0026] Step S102: performing a comparative analysis of settlement elements between the vector map data to be evaluated and the vector data of multiple time series, and obtaining the similarity of the number of settlement elements, the similarity of the total area of ​​settlement elements, and the geometric similarity of settlement elements between the vector map data to be evaluated and the vector data of each time series;

[0027] Step S103: setting the similarity weight of the number of settlement elements, the similarity weight of the total area of ​​settlement elements, and the geometric similarity weight of settlement elements, and calculating the weighted average score of the vector map data to be evaluated and the vector data of each time series based on the similarity of the number of settlement elements, the similarity weight of the number of settlement elements, the similarity weight of the total area of ​​settlement elements, the similarity weight of the total area of ​​settlement elements, the geometric similarity of settlement elements, and the geometric similarity weight of settlement elements;

[0028] Step S104: Analyze the weighted average score of the vector data of multiple time series to obtain the final evaluation result of the timeliness.

[0029] In specific implementation, the following steps are performed to collect multiple digital orthophoto data in the time series of the area to be evaluated:

[0030] The spatial range of the multiple digital orthophoto data in the time series is consistent with the spatial range of the vector map data to be evaluated; the multiple digital orthophoto data in the time series are evenly distributed in the time interval, and the image resolution is greater than the set resolution threshold, and the color saturation difference is within a certain threshold.

[0031] Specifically, collect multiple digital orthophotos over a time series for the area to be evaluated. Collect multiple digital orthophotos with a known time series. The area should be consistent with the spatial range of the vector map data to be evaluated, with evenly distributed time intervals, image resolution better than a resolution threshold, and similar color saturation. The resolution threshold can be set to 0.5 meters. Color saturation similarity can be determined using a saturation threshold.

[0032] In specific implementation, the following steps are used to extract residential features:

[0033] The settlement features of the digital orthophoto data are extracted using a method based on automatic sample expansion and / or a method based on deep learning is used to extract the settlement features of the digital orthophoto data and / or a method based on edge density features is used to extract the settlement features of the digital orthophoto data.

[0034] Specifically, extract the residential area elements. Figure 3 As shown, based on the artificial intelligence method, the residential features are automatically extracted from the prepared multi-time series (multiple images in a time series) digital orthophoto data, and multi-time series vector data with the same geometric accuracy as the original digital orthophoto data are generated.

[0035] In specific implementation, the following steps are performed to compare and analyze the residential elements of the vector map data to be evaluated with the vector data of multiple time series, and obtain the similarity of the number of residential elements, the similarity of the total area of ​​residential elements, and the geometric similarity of residential elements between the vector map data to be evaluated and the vector data of each time series:

[0036] Loop through the vector data of each time series in the vector data of multiple time series until the vector data of all time series are calculated: calculate the similarity of the number of settlement elements through the number of settlement elements in the vector map data to be evaluated and the number of settlement elements in the vector data of any time series; calculate the similarity of the total area of ​​settlement elements through the total area of ​​settlement elements in the vector map data to be evaluated and the total area of ​​settlement elements in the vector data of any time series; calculate the mean square error of geometric accuracy between the vector map data to be evaluated and the vector data of any time series, and calculate the geometric similarity of settlement elements through the mean square error of geometric accuracy.

[0037] Specifically, the vector map data is compared and analyzed. The vector map data to be evaluated is compared and analyzed with the multi-time series vector data generated in the previous step to obtain the similarity in the number of settlement elements, the similarity in the total area of ​​settlement elements, and the geometric similarity of settlement elements between the different time series vector data.

[0038] In specific implementation, the following steps are used to calculate the similarity of the number of residential features in the vector map data to be evaluated and the number of residential features in the vector data of any time series:

[0039] Similarity of the number of residential features Among them, m0 is the number of settlement elements in the vector map data to be evaluated, m is the number of settlement elements in the vector data of any time series, and p is the adjustment coefficient.

[0040] Specifically, the number of settlement elements represents the changes in settlements. The timeliness of the vector map data is determined by calculating and analyzing the proximity between the number of settlement elements in the data to be evaluated and the number of settlement elements in a certain time series vector data. The closer the number is to the number of settlement elements in a certain time series vector data, the more current the vector map data to be evaluated is considered to be. The similarity of the number of settlement elements can be calculated using the similarity calculation formula; the larger the result, the more similar it is.

[0041] The calculation formula for the similarity of the number of residential elements is: Among them, S i is the similarity analysis result of the number of residential elements, m0 is the number of residential elements in the vector map data to be evaluated, m is the number of residential elements in the comparison data, and p is the adjustment coefficient in the range of (0, 1). The larger p is, the stricter the restriction is.

[0042] As shown in Table 1, the examples of this application Figure 3 The similarity results of the number of residential elements of the vector map data can be obtained through experimental calculation:

[0043] Table 1 Similarity results of the number of residential elements in vector map data

[0044]

[0045] In specific implementation, the following steps are used to calculate the similarity of the total area of ​​residential features in the vector map data to be evaluated and the total area of ​​residential features in the vector data of any time series:

[0046] Similarity of total area of ​​residential elements Among them, m0 is the total area of ​​residential features in the vector map data to be evaluated, m is the total area of ​​residential features in the vector data of any time series, and p is the adjustment coefficient.

[0047] Specifically, the total area of ​​residential elements can also represent changes in residential areas. The timeliness of vector map data is determined by calculating and analyzing the degree of proximity between the total area of ​​residential elements of the data to be evaluated and the total area of ​​residential elements of a certain time series vector data. The closer the total area is to the total area of ​​residential elements of a certain time series vector data, the more current the vector map data to be evaluated is considered to be. By using the similarity calculation formula, it can be obtained that the larger the result, the more similar it is.

[0048] The calculation formula for the similarity of the total area of ​​residential elements is: Among them, S j is the similarity analysis result of the total area of ​​residential elements; m0 is the total area of ​​residential elements in the vector map data to be evaluated, m is the total area of ​​residential elements in the comparison data, and p is the adjustment coefficient in the range of (0, 1). The larger p is, the stricter the restriction is.

[0049] As shown in Table 2, the examples of this application Figure 3 The similarity results of the total area of ​​residential elements of the vector map data can be obtained through experimental calculation:

[0050] Table 2 Similarity results of total area of ​​residential features in vector map data

[0051]

[0052] In specific implementation, the following steps are used to calculate the mean square error of the geometric accuracy of the residential area features between the vector map data to be evaluated and the vector data of any time series, and the geometric similarity of the residential area features is calculated based on the mean square error of the geometric accuracy:

[0053] Geometric similarity of residential features Wherein, m0 is the mean error limit of the vector map data to be evaluated, and m is the mean error value of the geometric accuracy of the vector map data to be evaluated.

[0054] Specifically, based on advances in satellite positioning and Earth observation technologies, the more current the data, the smaller the geometric accuracy error and the higher the similarity. The currentness of the vector map data to be evaluated is determined by calculating the mean square error in geometric accuracy between the vector map data to be evaluated and a particular vector data item in the multi-time series vector data. The smaller the mean square error, the closer the currentness of the vector map data to be evaluated is to the currentness of the current time series vector data. In some embodiments, at a scale of 1:50,000, the mean square error limit does not exceed 25 meters.

[0055] The calculation formula for the geometric similarity of residential elements is:

[0056] Among them, S z is the geometric similarity result of residential features, m0 is the mean error limit of the vector map data to be evaluated, and m is the geometric accuracy mean error value of the vector map data to be evaluated.

[0057] As shown in Table 3, the examples of this application Figure 3 The geometric similarity results of residential area elements of vector map data can be obtained through experimental calculation:

[0058] Table 3 Geometric similarity results of residential area elements of vector map data

[0059]

[0060] In specific implementation, the following steps are performed to obtain the weighted average score of the vector map data to be evaluated and the vector data of each time series based on the similarity of the number of residential elements, the weight of the similarity of the number of residential elements, the similarity of the total area of ​​residential elements, the weight of the similarity of the total area of ​​residential elements, the geometric similarity of residential elements, and the weight of the geometric similarity of residential elements:

[0061] Calculate the weighted average scores of the vector map data to be evaluated and the vector data of each time series respectively until the weighted average scores of the vector data of all time series are calculated:

[0062] S n =∑(s i *p i , s j *p j , s z *p z ), where S n is the weighted average score of vector data of any time series, S i is the similarity of the number of residential elements, S j is the similarity of the total area of ​​residential elements, S z is the geometric similarity of residential elements, p i is the similarity weight of the number of residential elements, p j is the similarity weight of the total area of ​​residential elements, p z is the geometric similarity weight of residential features.

[0063] Specifically, the similarity in the number of settlement elements, the similarity in the total area of ​​settlement elements, and the geometric similarity of settlement elements between the vector map data to be evaluated and each time series vector data are statistically analyzed. By setting the weights for the similarity in the number of settlement elements, the weights for the similarity in the total area of ​​settlement elements, and the weights for the geometric similarity of settlement elements to (0.3, 0.5, 0.2), respectively, the weighted average method is used for calculation and analysis to obtain the weighted average score of the vector map data to be evaluated and each multi-time series vector data.

[0064] The weighted average formula is: S n =∑(s i *p i , s j *p j , s z *p z ), where S n is the weighted average score, S i ,S j ,S z is the score of single item analysis, p i ,p j ,p z is the weight of the corresponding similarity.

[0065] As shown in Table 4, the examples of this application Figure 3 The weighted average results of the similarity of the number of settlement elements, the total area similarity of settlement elements, and the geometric similarity of settlement elements between the vector data to be evaluated can be obtained through experimental calculation:

[0066] Table 4 Similarity weighted average results

[0067]

[0068] In specific implementation, the following steps are performed to obtain the final evaluation result of the current situation based on the weighted average result analysis of the vector data of multiple time series:

[0069] Obtain the maximum value of the weighted average score and the second maximum value of the weighted average score in the vector data of multiple time series; obtain the first time point corresponding to the maximum value of the weighted average score, obtain the second time point corresponding to the second maximum value of the weighted average score, and generate the time series interval corresponding to timeliness through the first time point and the second time point; use the time series interval corresponding to timeliness as the final evaluation result of timeliness.

[0070] In specific implementation, the following steps are performed to obtain the maximum value and the second maximum value of the weighted average score in the vector data of multiple time series:

[0071] S=max(S n )(n=1,2,...,n), where S is the maximum and second maximum value, Sn is the weighted average score of the nth data, n is the time series number, and max is the function of taking the maximum value and the second largest value.

[0072] Specifically, based on the weighted average scores of the vector map data to be evaluated and all time series vector data, the maximum and second largest values ​​of the results are taken, and the time series corresponding to the corresponding vector data are used to determine in which two time intervals the current nature of the vector map data to be evaluated falls.

[0073] The formula for taking the maximum and second maximum values ​​is: S=max(S n )(n=1,2,...,n), where S is the maximum or second largest value, S n is the weighted average score of the nth data, and max is the maximum and second largest values.

[0074] The maximum value calculated in the previous step is 63.83560473, and the second largest value is 54.18363162, so the timeliness of the data to be evaluated should be from 2021 to 2024.

[0075] like Figure 4 As shown, through the overlay analysis and verification of the vector data to be evaluated and the digital orthophoto data of known time series, the verification results meet the following requirements: the timeliness of the vector map data to be evaluated is closest to the distribution of residential elements in the digital orthophoto data from 2021 to 2024.

[0076] The embodiments of the present invention achieve the following technical effects:

[0077] The vector map data timeliness assessment method of the embodiment of the present invention utilizes remote sensing image feature extraction technology, combined with statistical analysis methods and quality assessment methods to perform vector data timeliness assessment. Compared with traditional vector data timeliness assessment methods, it improves the accuracy of timeliness assessment, solves the problems of low assessment efficiency, inability to quantify assessment results, and difficulty in coping with large-scale applications; it integrates the settlement feature extraction capabilities of artificial intelligence and traditional mathematical processing algorithms, and fully combines their respective advantages; by extracting settlement features through artificial intelligence methods, it can quickly realize the establishment of high-precision multi-time series vector data, and realize automatic comparison, thereby improving processing efficiency; by using relevant quality evaluation methods to quantify the assessment results, it avoids the human subjective factors in data assessment, thereby improving the accuracy of data assessment.

[0078] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for assessing the currentness of vector map data based on multi-temporal remote sensing images, characterized in that: include: Acquire multiple digital orthophoto image data in a time series, extract residential area features from the multiple digital orthophoto image data in the time series respectively, and generate multi-time series vector data, wherein the multi-time series vector data is consistent with the spatial range of the vector map data to be evaluated, and the multi-time series vector data includes multiple single-time series vector data; Comparing and analyzing the residential area elements of the vector map data to be evaluated and the vector data of the multiple time series to obtain the similarity of the number of residential area elements, the similarity of the total area of ​​residential area elements, and the geometric similarity of residential area elements between the vector map data to be evaluated and the vector data of each time series, including: Loop through the vector data of each time series in the multiple time series vector data until the vector data of all time series are calculated: The similarity of the number of residential elements is calculated by the number of residential elements in the vector map data to be evaluated and the number of residential elements in any time series vector data. ,in, is the number of residential area elements in the vector map data to be evaluated, is the number of residential features in any time series vector data, is the adjustment factor; The similarity of the total area of ​​residential elements is calculated by the total area of ​​residential elements in the vector map data to be evaluated and the total area of ​​residential elements in any time series vector data. ,in, is the total area of ​​the residential area elements in the vector map data to be evaluated, is the total area of ​​residential features of vector data of any time series, is the adjustment factor; Calculate the geometric accuracy mean square error between the vector map data to be evaluated and the vector data of any time series, and calculate the geometric similarity of the residential elements through the geometric accuracy mean square error. ,in, S z is the geometric similarity of residential features, is the mean error limit, is the mean error of geometric accuracy; Setting a similarity weight for the number of settlement elements, a similarity weight for the total area of ​​settlement elements, and a geometric similarity weight for settlement elements, and calculating a weighted average score result of the vector map data to be evaluated and the vector data of each time series based on the similarity weight for the number of settlement elements, the similarity weight for the number of settlement elements, the similarity weight for the total area of ​​settlement elements, the similarity weight for the total area of ​​settlement elements, the geometric similarity weight for settlement elements, and the geometric similarity weight for settlement elements; The final evaluation result of the timeliness is obtained based on the weighted average score analysis of the vector data of the multiple time series, including: Obtaining the maximum value of the weighted average score and the second maximum value of the weighted average score in the vector data of multiple time series; Obtaining a first time point corresponding to the maximum value of the weighted average score, obtaining a second time point corresponding to the second maximum value of the weighted average score, and generating a time series interval corresponding to currentness through the first time point and the second time point; The time series interval corresponding to the timeliness is used as the final evaluation result of the timeliness.

2. The method for assessing the currentness of vector map data based on multi-temporal remote sensing images according to claim 1, wherein: The weighted average score of the vector map data to be evaluated and the vector data of each time series is calculated based on the similarity of the number of settlement elements, the weight of the similarity of the number of settlement elements, the similarity of the total area of ​​settlement elements, the weight of the similarity of the total area of ​​settlement elements, the geometric similarity of settlement elements, and the geometric similarity weight of settlement elements, including: Calculate the weighted average scores of the vector map data to be evaluated and the vector data of each time series respectively until the weighted average scores of the vector data of all time series are calculated: ,in, is the weighted average score of vector data of any time series, is the similarity of the number of residential elements, is the similarity of the total area of ​​the residential elements, is the geometric similarity of the residential elements, is the similarity weight of the number of residential elements, is the similarity weight of the total area of ​​the residential area element, is the geometric similarity weight of the residential area element.

3. The method for assessing the currentness of vector map data based on multi-temporal remote sensing images according to claim 1, wherein: Obtaining the maximum value of the weighted average score and the second maximum value of the weighted average score in the vector data of multiple time series, including: ,in, is the maximum or second largest value, For the The weighted average score of the data, is the time series number, is a function that takes the maximum and second largest values.

4. The method for assessing the currentness of vector map data based on multi-temporal remote sensing images according to claim 1, wherein: The plurality of digital orthophoto data in the time series are consistent with the spatial range of the vector map data to be evaluated; The multiple digital orthophoto data in the time series are evenly distributed in time intervals, and the image resolution is greater than a set resolution threshold, and the color saturation difference is within a saturation threshold.

5. The method for assessing the currentness of vector map data based on multi-temporal remote sensing images according to claim 1, wherein: Also includes: Extracting the residential features of the digital orthophoto data using a method based on automatic sample expansion and / or, Extracting the residential features from the digital orthophoto data using a deep learning-based method and / or, The residential area elements of the digital orthophoto data are extracted using a method based on edge density features.

Images