A method for intelligent analysis of geographic information of a remote sensing mapping image

Abstract

The present application relates to the technical field of remote sensing image recognition, and particularly relates to a kind of geographic information intelligent analysis method of remote sensing mapping image.The method identifies the ridge region in the mountain remote sensing mapping image using ridge region growing.In the ridge region growing process, the first ridge gray scale feature of the neighborhood point is obtained according to the gray scale distribution characteristics of the mountain to the light surface and the back light surface.The second gray scale feature is obtained according to the change trend of the first ridge gray scale feature in the growing process.The adjustment weight under different directions is obtained according to the growth direction distribution of the growing point in the growing process, and the second gray scale feature is adjusted to obtain the third gray scale feature.The growing point is determined according to the third gray scale feature.The present application improves the region growing rule based on the characteristics of the ridge, and accurately identifies the ridge region in the mountain remote sensing image.

Description

Technical Field

[0001] This invention relates to the field of remote sensing image recognition technology, and more specifically to a method for intelligent analysis of geographic information from remote sensing and mapping images. Background Technology

[0002] Remote sensing imagery is image data of the Earth's surface acquired through remote sensing technology. It contains a wealth of geospatial information. By analyzing remote sensing images, the distribution of terrain, landforms, and features can be accurately identified, thereby assisting researchers in conducting geographical studies and other work in the region. In the process of remote sensing imagery recognition, the identification and labeling of ridge areas is of great significance, as ridge areas provide a more intuitive and detailed understanding of the terrain's undulations and landform changes.

[0003] In existing technologies, ridge regions can be obtained by performing region growing on remote sensing images. However, because the ridge lines in remote sensing images are complex, and because the pixel features of the ridges are similar to those of the shadowed side of the mountain, the region growing process based on a fixed grayscale threshold can result in the obtained ridge regions containing a large amount of information about the shadowed side of the mountain, leading to unclear ridge region identification. Summary of the Invention

[0004] To address the technical problem of unclear ridge region identification caused by existing region growing algorithms that grow ridge regions based on grayscale thresholds, the present invention aims to provide an intelligent geographic information analysis method for remote sensing mapping images. The specific technical solution adopted is as follows:

[0005] This invention proposes an intelligent analysis method for geographic information from remote sensing mapping images, the method comprising:

[0006] Acquire a remote sensing image of the mountain; perform ridge region growth on the remote sensing image of the mountain based on preset seed points to obtain a ridge region; the method for ridge region growth includes:

[0007] In the process of determining whether a neighboring point is a new potential growth point, the direction between the potential growth point and its neighboring points is taken as the growth direction. Based on the difference in grayscale distribution characteristics between preset neighboring regions on both sides of the growth direction, the first ridge grayscale feature of the neighboring point is obtained. Based on the changing trend of the first ridge grayscale feature between the previously grown points and the neighboring points, and the difference in the first ridge grayscale feature between the neighboring points and the potential growth point, the second ridge grayscale feature of the neighboring point is obtained. Based on the growth direction distribution characteristics between the already grown points and the rate of change of the growth direction, the adjustment weights of the potential growth point in different directions are obtained. The second ridge grayscale feature is adjusted according to the adjustment weights corresponding to the growth direction to obtain the third ridge grayscale feature. The third ridge grayscale feature is used to determine whether the neighboring point is a new potential growth point.

[0008] Furthermore, the method for obtaining the grayscale features of the first ridge includes:

[0009] Obtain the overall grayscale feature differences between the two neighboring regions;

[0010] Obtain the difference in grayscale distribution between the two adjacent regions;

[0011] The product of the overall grayscale feature difference and the grayscale distribution difference is normalized to obtain the grayscale feature of the first ridge.

[0012] Furthermore, the method for obtaining the overall grayscale feature difference includes:

[0013] The average gray value difference between the two neighboring regions is obtained, and the ratio of the average gray value difference to the minimum average gray value in the two neighboring regions is taken as the overall gray value feature difference.

[0014] Furthermore, the method for obtaining the grayscale distribution difference includes:

[0015] The difference in variance between the two neighboring regions is taken as the difference in gray-scale distribution.

[0016] Furthermore, the method for obtaining the grayscale features of the second ridge includes:

[0017] The grown points and their neighboring points are sequentially numbered according to their growth order, and the numbers are arranged in a fixed order to obtain a number sequence. The grayscale features of the first ridge corresponding to the number in the number sequence constitute a first ridge grayscale feature sequence. The correlation coefficient between the number sequence and the first ridge grayscale feature sequence is obtained, and the correlation coefficient is multiplied by the number of elements in the number sequence to obtain the growth characteristics. Based on the growth characteristics, the difference in grayscale features of the first ridge, and the grayscale features of the first ridge of the neighboring points, the second ridge grayscale features of the neighboring points are obtained. The grayscale features of the first ridge and the second ridge are positively correlated, and the difference in grayscale features between the growth characteristics and the first ridge is negatively correlated with the grayscale features of the second ridge.

[0018] Furthermore, the method for obtaining the adjustment weights includes:

[0019] The main direction of the ridge is determined based on the growth direction between adjacent growth points.

[0020] The adjustment weights include the second weight along the main ridge direction and the first weight along the non-main ridge direction;

[0021] The directional weight is obtained based on the frequency of the main ridge direction among all growth directions; the first weight of the point to be grown in a non-main ridge direction is obtained based on the directional weight and the rate of change; the second weight is obtained based on the first weight, and the second weight is negatively correlated with the first weight.

[0022] Furthermore, the main direction of the ridge is the growth direction that appears most frequently between adjacent growth points.

[0023] Furthermore, the formula for obtaining the first weight includes:

[0024] Where P is the first weight of the growth point, exp() is an exponential function with the natural constant as the base, p is the frequency of the main direction of the ridge, norm{} is the normalization function, L is the number of grown points, and G ′ i G represents the grayscale feature of the second ridge at the i-th grown point. ′ i-1 For the grayscale feature of the second ridge at the (i-1)th grown point, k i Let k be the slope of the growth direction from the i-th grown point to the (i+1)-th grown point. i-1 It is the slope of the growth direction from the (i-1)th grown point to the ith grown point.

[0025] Furthermore, the method for obtaining the grayscale features of the third ridge includes:

[0026] The product of the adjusted weight and the second ridge grayscale feature is used as the third ridge grayscale feature.

[0027] Furthermore, determining whether the neighboring points are new growth points based on the grayscale features of the third ridge includes:

[0028] If the grayscale feature of the third ridge is greater than the preset growth threshold, then the neighboring point is taken as a new growth point.

[0029] The present invention has the following beneficial effects:

[0030] In this embodiment of the invention, during the growth process in the ridge region, it is considered that the ridge is located between the shaded and illuminated sides of the mountain, and there is a significant difference in grayscale distribution between these two areas. Therefore, in determining new growth points, the grayscale characteristics of the first ridge are analyzed. Furthermore, considering that during the growth process, the grayscale characteristics of the first ridge become less pronounced near the ridge endpoint due to the undulations of the mountain, the trend of the grayscale characteristics of the first ridge is correlated with the growth process of the growth point. Based on this, the grayscale characteristics of the second ridge are obtained. Further considering that there may be chaotic edge trends in the ridge endpoint area, the changes and distribution of the growth direction during the growth process are analyzed to determine the grayscale characteristics of the third ridge. Based on the grayscale characteristics of the third ridge, an accurate new growth point can be determined. The ridge region can be effectively identified through the ridge region growth process. Attached Figure Description

[0031] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A flowchart illustrating a method for intelligent analysis of geographic information from remote sensing images, provided in one embodiment of the present invention.

[0033] Figure 2 A schematic diagram of a mountain remote sensing image provided in one embodiment of the present invention;

[0034] Figure 3 This is a flowchart illustrating a method for growing ridge regions according to an embodiment of the present invention;

[0035] Figure 4 This is a schematic diagram of grayscale feature analysis of a first ridge provided in an embodiment of the present invention. Detailed Implementation

[0036] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a geographic information intelligent analysis method for remote sensing mapping images proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0038] The specific scheme of the intelligent analysis method for geographic information of remote sensing mapping images provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0039] Please see Figure 1 The diagram illustrates a flowchart of a method for intelligent analysis of geographic information from remote sensing images according to an embodiment of the present invention. The method includes:

[0040] Step S101: Obtain remote sensing images of the mountain.

[0041] In this embodiment of the invention, the mountain remote sensing image is obtained by capturing images of the mountain area using a remote sensing satellite or a camera-equipped UAV remote sensing platform, thus acquiring clear mountain remote sensing images for ridge identification. It should be noted that the initial mountain remote sensing image is a color image. To enhance image quality and facilitate image recognition, preprocessing steps such as atmospheric correction, geometric correction, and grayscale conversion are required to obtain the final mountain remote sensing image. Please refer to [link to relevant documentation]. Figure 2 This illustration shows a schematic diagram of a mountain remote sensing image provided by an embodiment of the present invention. It should be noted that image preprocessing techniques such as atmospheric correction, geometric correction, and grayscale conversion of remote sensing images are well-known to those skilled in the art and will not be elaborated upon here.

[0042] Step S102: In the remote sensing mapping image of the mountain, the ridge region is grown according to the preset seed points to obtain the ridge region.

[0043] In this embodiment of the invention, seed points are uniformly set in the remote sensing image of the mountain for ridge region growth. The specific number of seed points can be set according to the specific image resolution and is not limited here. In this embodiment of the invention, two hundred seed points are set for ridge region growth.

[0044] Please see Figure 3The diagram illustrates a flowchart of a method for growing ridge regions according to an embodiment of the present invention, the method comprising:

[0045] Step S1: In the process of determining whether the neighboring points of the point to be grown are new points to be grown, the direction between the point to be grown and the neighboring points is taken as the growth direction; based on the difference in gray-scale distribution characteristics between the preset neighboring regions on both sides of the growth direction, the gray-scale characteristics of the first ridge of the neighboring point are obtained.

[0046] In the process of growing the ridge region, similar to the conventional region growing algorithm, each new growth point is obtained and used as a point to be grown. Then, new points to be grown are searched for in the neighborhood points within the eight-neighborhood of the point to be grown. That is, the region growing process can be regarded as the process of whether the neighborhood points of the point to be grown meet the growth conditions.

[0047] It should be noted that, in another embodiment of the present invention, to make the ridge region growth more accurate, after determining the neighboring points within the eight-neighbor range of the point to be grown, the conventional region growing algorithm uses a gray-level difference threshold method to filter out the neighboring points participating in subsequent calculations. The remaining neighboring points are considered unnecessary to participate in subsequent calculations, thus reducing the computational load while increasing the accuracy of the final result. In one embodiment of the present invention, neighboring points with a gray-level difference value less than 5 from the point to be grown are used as neighboring points in the subsequent analysis process.

[0048] like Figure 2 As shown, a ridge is a prominent part of a mountain range or between mountain ridges. It has a relatively high altitude and steep slopes on both sides. Therefore, during the day, sunlight will create obvious light-facing and shadow-facing sides. There is a significant difference in grayscale distribution between the light-facing and shadow-facing sides. Therefore, if a neighboring point is a pixel in the ridge region, the direction between the neighboring point and the corresponding growth point, i.e., the growth direction, should be the direction of the ridge trend. The difference in grayscale distribution between the preset neighboring regions on both sides of the growth direction will follow the difference in grayscale distribution between the shadow-facing and light-facing sides. Therefore, the first ridge grayscale feature of the neighboring point can be obtained based on the difference in grayscale distribution between the preset neighboring regions on both sides of the growth direction. That is, the larger the first ridge grayscale feature, the more the neighboring point belongs to the ridge region pixel, and the more it should be a new growth point.

[0049] In one embodiment of the present invention, two 3×3 regions symmetrically positioned around a neighboring point on either side of the growth direction are designated as preset neighboring regions on both sides. Please refer to [link to relevant documentation]. Figure 4 The diagram illustrates a first ridge grayscale feature analysis according to an embodiment of the present invention, wherein x1 is the point to be grown, x2 is a neighboring point of the point to be grown, and two 3×3 regions symmetrical about x2 and on both sides of the direction from x1 to x2 are the preset neighboring regions.

[0050] Preferably, in one embodiment of the present invention, considering that the illuminated side is brighter and the shaded side is darker in a remote sensing image of a mountain, i.e., the illuminated side has a larger overall grayscale value than the shaded side, the overall grayscale feature difference between the two neighboring regions is first obtained. Further considering that the illuminated side has clearer image information and more mountain surface texture information due to the influence of illumination, the grayscale distribution of the illuminated side is more chaotic; while the shaded side, due to its higher altitude, has a lower grayscale value in the shadow formed by the shaded side of the ridge compared to the shadows in other areas of the image, thus the shaded side does not have obvious mountain surface texture features and the grayscale distribution is more uniform, therefore the grayscale distribution difference between the two neighboring regions can be further obtained. Combining the overall grayscale feature difference and the grayscale distribution difference between the two neighboring regions, the product of the overall grayscale feature difference and the grayscale distribution difference is normalized to obtain the first ridge grayscale feature. The first ridge feature is expressed by the formula:

[0051] G = norm{A×B}; where G is the first ridge feature, A is the overall gray-level feature difference, B is the gray-level distribution difference, and norm{} is the normalization function. In one embodiment of the present invention, the normalization function adopts the linear normalization method. In other embodiments of the present invention, range standardization or other function mapping methods can also be used to achieve normalization, which are all well-known technical means in the art and will not be described in detail here.

[0052] Preferably, in one embodiment of the present invention, the method for obtaining the overall grayscale feature difference includes:

[0053] The average grayscale difference between the two neighboring regions is obtained. A larger average grayscale difference indicates that the two neighboring regions are closer to the grayscale distribution characteristics of the lit and shaded sides. Because the shaded side is darker than other shadows in the image, the ratio of the average grayscale difference to the minimum average grayscale value in the two neighboring regions is taken as the overall grayscale feature difference. That is, the smaller the minimum average grayscale value and the larger the average grayscale difference, the closer the two neighboring regions are to the grayscale distribution characteristics between the lit and shaded sides. The overall grayscale feature difference is expressed by the formula:

[0054] in, This represents the average gray value of one of the neighboring regions on either side. The average gray value of the neighboring region on the other side is given by , and min() is the minimum value selection function.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining grayscale distribution differences includes:

[0056] The difference in variance between the two adjacent regions is taken as the difference in gray-level distribution. For the illuminated side, because it contains more information about the mountain texture, the gray-level variance of the illuminated side region should be larger; for the shaded side, because it is generally darker and contains less information, the gray-level distribution is more uniform, and the variance should be smaller. Therefore, the larger the difference in variance, the closer the two adjacent regions are to the gray-level distribution characteristics between the illuminated and shaded sides. The difference in gray-level distribution is expressed by the formula:

[0057] B = |σ(g1) - σ(g2)|; where σ(g1) is the variance of one of the neighboring regions and σ(g2) is the variance of the other neighboring region.

[0058] Step S2: Based on the changing trend of the grayscale features of the first ridge between the grown points and the neighboring points, and the difference in grayscale features of the first ridge between the neighboring points and the points to be grown, obtain the grayscale features of the second ridge of the neighboring points.

[0059] Considering the complexity of information in remote sensing images of mountains, if growth is based solely on the grayscale features of the first ridge, noise points during the growth process might be mistaken for pixels within the ridge region. Furthermore, given the distinct edge trends of ridges, the grayscale features of adjacent growth points within the ridge region should exhibit significant similarity. Moreover, as the ridge edge extends, the grayscale features at the ridge's endpoint (the ridge's end) gradually become blurred, and the corresponding first ridge grayscale feature value should gradually weaken. Therefore, as the growth process progresses and the growth point strengthens, its corresponding first ridge grayscale feature value should exhibit a consistent weakening trend. Thus, by analyzing neighboring points and previously grown points, the changing trend of the first ridge grayscale features can be determined. Further analysis of the differences in first ridge grayscale features between neighboring points and the point to be grown yields the second ridge grayscale features of the neighboring points. It should be noted that since the point to be grown is also a determined growth point, it is considered part of the previously grown points.

[0060] Preferably, in one embodiment of the present invention, the method for obtaining the grayscale features of the second ridge includes:

[0061] The grown points and their neighboring points are sequentially numbered according to their growth order, and these numbers are arranged in a fixed order to obtain a number sequence. In one embodiment of the present invention, starting from the first grown point from the seed point, it is numbered 1, the next grown point from the seed point is numbered 2, and so on, with the neighboring points numbered N, that is, the number sequence is {1,2,3…N}.

[0062] The gray-level features of the first ridge corresponding to the label in the label sequence constitute the first ridge gray-level feature sequence, that is, the length of the first ridge gray-level feature sequence is the same as that of the label sequence. In one embodiment of the present invention, the first ridge gray-level feature sequence corresponding to the label sequence is {G1, G2, G3…G}. N}, where G1 represents the grayscale feature of the first ridge corresponding to the first growth point, and so on, G N This represents the grayscale feature of the first ridge in the neighborhood.

[0063] Because the grayscale features of the ridge at the end of the ridge region gradually become blurred, the corresponding grayscale features of the first ridge will weaken, meaning the grayscale feature sequence of the first ridge should show a decaying trend. However, the label sequence is linearly increasing. Therefore, if the neighboring points are normal ridge region pixels, the grayscale feature sequence of the first ridge should have a relatively significant negative correlation with the label sequence. Thus, the correlation coefficient between the label sequence and the grayscale feature sequence of the first ridge is obtained. Further multiplying the correlation coefficient by the number of elements in the label sequence yields the growth characteristics; the more growth points there are, the closer to the end of the ridge, and the more important the relevant information represented by the correlation coefficient. Based on the growth characteristics, the difference in grayscale features of the first ridge, and the grayscale features of the neighboring points' first ridge, the grayscale features of the neighboring points' second ridge are obtained. The grayscale features of the first ridge and the second ridge are positively correlated, while the growth characteristics and the difference in grayscale features of the first ridge are negatively correlated with the grayscale features of the second ridge. The smaller the growth characteristic, the more the correlation information between the gray-scale feature sequence of the first ridge and the label sequence conforms to the normal trend of ridge feature changes; the smaller the difference in gray-scale features of the first ridge, the more similar the gray-scale features of the neighboring point and the point to be grown are, and the greater the probability that the neighboring point is the next growth point of the point to be grown.

[0064] In one embodiment of the present invention, the grayscale feature of the second ridge is expressed by the formula:

[0065] G ′ = G×exp{-|G-G0|}×exp{-M×r}; where G ′ Let G be the gray-scale feature of the second ridge of the neighboring points, G be the gray-scale feature of the first ridge of the neighboring points, G0 be the gray-scale feature of the first ridge of the point to be grown, exp{} be the exponential function with the natural constant as the base, M be the number of elements in the label sequence, and r be the correlation coefficient between the label sequence and the gray-scale feature sequence of the first ridge.

[0066] In the formula for the grayscale feature of the second ridge, the data is negatively correlated and normalized by an exponential function with the natural constant as the base. The smaller |G-G0| is, the more similar the grayscale features of the first ridge are between the neighboring points and the point to be grown, and the more likely the neighboring points are to be the next growth points of the point to be grown. The smaller M×r is, the closer the correlation between the label sequence and the grayscale feature sequence of the first ridge is to the correlation of the normal ridge area, and the more likely the neighboring points are to be the next growth points of the point to be grown. The larger G is, the more the neighboring points match the grayscale features of the pixels in the ridge area, and the more likely the neighboring points are to be the next growth points of the point to be grown.

[0067] In one embodiment of the present invention, the correlation coefficient is selected from the Pearson correlation coefficient. The larger the Pearson correlation coefficient, the more correlated the two sequences are. Considering that the Pearson correlation coefficient can have negative values, in order to facilitate calculation, a positive integer 1 is added after obtaining the Pearson correlation coefficient to obtain the correlation coefficient.

[0068] Step S3: Based on the distribution characteristics of growth directions among the grown points and the rate of change of growth directions, obtain the adjustment weights of the points to be grown in different directions; adjust the grayscale features of the second ridge according to the adjustment weights corresponding to the directions to be grown, and obtain the grayscale features of the third ridge; determine whether the neighboring points are new points to be grown based on the grayscale features of the third ridge.

[0069] Because remote sensing images have a wide acquisition range and contain numerous elements, including roads, rivers, houses, and cables, these elements may exhibit grayscale features similar to those of mountain ridges due to their shapes and structures. Therefore, in addition to grayscale analysis, it is necessary to consider the significant extension characteristics of mountain ridges. Along the entire ridgeline, higher elevations typically exhibit a relatively straight shape with less variation in shape and frequency during extension. However, at the ridge's end, due to the change in elevation, the information becomes more complex, and shape variations increase. This means that while the grayscale features of the ridge weaken at the end, the abrupt change in shape becomes more pronounced. Therefore, this invention analyzes the distribution of growth directions between adjacent grown points and the rate of change of these directions. Based on the characteristics of the growth directions, the location of the point to be grown is determined, and different weights are assigned to different directions based on its location. Adjusting the weights according to the direction to be grown allows for the adjustment of the grayscale features of the second ridge of neighboring points, resulting in the final grayscale features of the third ridge used for evaluation.

[0070] Preferably, in one embodiment of the present invention, considering that the edges of the ridge generally tend to follow a single dominant direction, the directions are categorized into two types: the dominant ridge direction and non-dominant ridge directions. The dominant ridge direction can be determined based on the growth directions between adjacent grown points. In one embodiment of the present invention, the dominant ridge direction is the growth direction with the highest frequency of occurrence between adjacent grown points. In other embodiments of the present invention, data analysis methods such as principal component analysis can also be used to determine the final dominant direction. The specific technical means are well-known to those skilled in the art and will not be elaborated here.

[0071] Furthermore, the directional weight can be obtained based on the frequency of the ridge's main direction among all growth directions. A higher frequency indicates that most of the growth pattern in the currently grown area follows the ridge's main direction, meaning the growth point is located in the middle of the ridge region and should better conform to the ridge's main direction. The first weight for non-ridge main directions should be reduced. Conversely, a lower frequency indicates the growth point is located at the ridge's end, which better conforms to the non-ridge main direction, and the first weight for non-ridge main directions should be increased. A higher rate of change in growth direction indicates the growth point is closer to the ridge's end, and the higher the first weight should be. Therefore, the first weight of the growth point in non-ridge main directions can be obtained based on the directional weight and rate of change. The second weight in the ridge main direction can then be obtained based on the first weight. Since there are only two types of directions, the second weight is negatively correlated with the first weight.

[0072] Preferably, in one embodiment of the present invention, the formula for obtaining the first weight includes:

[0073] Where P is the first weight of the growth point, exp() is the exponential function with the natural constant as the base, p is the frequency of the main ridge direction, norm{} is the normalization function, L is the number of grown points, and G ′ i G represents the grayscale feature of the second ridge at the i-th grown point. ′ i-1 For the grayscale feature of the second ridge at the (i-1)th grown point, k i Let k be the slope of the growth direction from the i-th grown point to the (i+1)-th grown point. i-1 It is the slope of the growth direction from the (i-1)th grown point to the ith grown point.

[0074] In the formula for obtaining the first weight, the frequency of the main ridge direction is first negatively correlated and normalized using an exponential function with the natural constant as the base; that is, the higher the frequency, the smaller the first weight. Furthermore, the rate of change is obtained by utilizing the difference in the slope of the growth direction between adjacent grown points. The rate of change is the reciprocal of the average grayscale characteristic of the second ridge between two adjacent growth points. As a weight, The larger the rate of change, the more important the slope difference becomes, indicating that the two existing growth points are located at the end of the ridge. In this case, a larger rate of change means the growth point is closer to the ridge end, making it easier to extend in directions other than the main ridge direction. Therefore, the first weight should be larger. Conversely, if the frequency is high and the rate of change is low, it indicates the growth point is located in the middle of the ridge edge, making it easier to extend along the main ridge direction. Therefore, the first weight should be smaller, and the second weight should be larger.

[0075] It should be noted that, in one embodiment of the present invention, the value range of the first weight is between 0 and 1. Therefore, the second weight is obtained by subtracting the first weight from the positive integer 1. That is, the sum of the second weight and the first weight is a positive integer 1.

[0076] In one embodiment of the present invention, the adjustment weight corresponding to the growth direction is determined, and then the product of the adjustment weight and the grayscale feature of the second ridge is used as the grayscale feature of the third ridge.

[0077] In one embodiment of the present invention, a growth threshold is set. If the grayscale feature of the third ridge is greater than the preset growth threshold, then the neighboring points are taken as new growth points. In another embodiment of the present invention, considering the result of uniform normalization of various indicators, the grayscale feature of the third ridge is also the result of normalization, therefore the growth threshold is set to 0.4.

[0078] By continuously performing the ridge region growth process on each seed point until the growth is complete, multiple region growth results can be obtained. In this embodiment of the invention, incomplete growth results only need to be removed. These incomplete growth results are caused by the seed point not being a ridge pixel. Retaining the fully grown results yields the ridge region. Marking the ridge region enables intelligent analysis of the mountain. It should be noted that a growth region area threshold can be set. Growth results smaller than the growth region area threshold are considered incomplete growth results and are removed. The growth region area threshold can be set according to the image resolution and field of view, and is not limited here. In other embodiments of the invention, conventional techniques from other region growth algorithms can also be used to determine incomplete growth results, which will not be elaborated or limited here.

[0079] In summary, this invention utilizes ridge region growing to identify ridge regions in remote sensing images of mountains. During the ridge region growing process, a first ridge gray-scale feature is obtained based on the gray-scale distribution characteristics of the illuminated and shaded sides of the mountain. A second gray-scale feature is obtained based on the changing trend of the first ridge gray-scale feature during the growing process. Adjustment weights are obtained for different directions based on the growth direction distribution of growth points during the growing process, and the second gray-scale feature is adjusted to obtain a third gray-scale feature. The growth point is determined based on the third gray-scale feature. This invention improves the region growing rule based on ridge features, enabling accurate identification of ridge regions in remote sensing images of mountains.

[0080] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for intelligent analysis of geographic information from remote sensing and mapping images, characterized in that, The method includes: Acquire a remote sensing image of the mountain; perform ridge region growth on the remote sensing image of the mountain based on preset seed points to obtain a ridge region; the method for ridge region growth includes: In the process of determining whether a neighboring point is a new potential growth point, the direction between the potential growth point and its neighboring points is taken as the growth direction. Based on the difference in grayscale distribution characteristics between preset neighboring regions on both sides of the growth direction, the first ridge grayscale feature of the neighboring point is obtained. Based on the changing trend of the first ridge grayscale feature between the previously grown points and the neighboring points, and the difference in the first ridge grayscale feature between the neighboring points and the potential growth point, the second ridge grayscale feature of the neighboring point is obtained. Based on the growth direction distribution characteristics between the already grown points and the rate of change of the growth direction, the adjustment weights of the potential growth point in different directions are obtained. The second ridge grayscale feature is adjusted according to the adjustment weights corresponding to the growth direction to obtain the third ridge grayscale feature. The third ridge grayscale feature is used to determine whether the neighboring point is a new potential growth point.

2. The method for intelligent analysis of geographic information from remote sensing mapping images according to claim 1, characterized in that, The method for obtaining the grayscale features of the first ridge includes: Obtain the overall grayscale feature differences between the two neighboring regions; Obtain the difference in grayscale distribution between the two adjacent regions; The product of the overall grayscale feature difference and the grayscale distribution difference is normalized to obtain the grayscale feature of the first ridge.

3. The method for intelligent analysis of geographic information from remote sensing images according to claim 2, characterized in that, The method for obtaining the overall grayscale feature difference includes: The average gray value difference between the two neighboring regions is obtained, and the ratio of the average gray value difference to the minimum average gray value in the two neighboring regions is taken as the overall gray value feature difference.

4. The method for intelligent analysis of geographic information from remote sensing and mapping images according to claim 2, characterized in that, The method for obtaining the grayscale distribution difference includes: The difference in variance between the two neighboring regions is taken as the difference in gray-scale distribution.

5. The method for intelligent analysis of geographic information from remote sensing mapping images according to claim 1, characterized in that, The methods for obtaining the grayscale features of the second ridge include: The grown points and their neighboring points are sequentially numbered according to their growth order, and the numbers are arranged in a fixed order to obtain a number sequence. The grayscale features of the first ridge corresponding to the number in the number sequence constitute a first ridge grayscale feature sequence. The correlation coefficient between the number sequence and the first ridge grayscale feature sequence is obtained, and the correlation coefficient is multiplied by the number of elements in the number sequence to obtain the growth characteristics. Based on the growth characteristics, the difference in grayscale features of the first ridge, and the grayscale features of the first ridge of the neighboring points, the second ridge grayscale features of the neighboring points are obtained. The grayscale features of the first ridge and the second ridge are positively correlated, and the difference in grayscale features between the growth characteristics and the first ridge is negatively correlated with the grayscale features of the second ridge.

6. The method for intelligent analysis of geographic information from remote sensing mapping images according to claim 1, characterized in that, The method for obtaining the adjustment weights includes: The main direction of the ridge is determined based on the growth direction between adjacent growth points. The adjustment weights include the second weight along the main ridge direction and the first weight along the non-main ridge direction; The directional weight is obtained based on the frequency of the main ridge direction among all growth directions; the first weight of the point to be grown in a non-main ridge direction is obtained based on the directional weight and the rate of change; the second weight is obtained based on the first weight, and the second weight is negatively correlated with the first weight.

7. The method for intelligent analysis of geographic information from remote sensing mapping images according to claim 6, characterized in that, The main direction of the ridge is the growth direction that appears most frequently between adjacent growth points.

8. The method for intelligent analysis of geographic information from remote sensing and mapping images according to claim 6, characterized in that, The formula for obtaining the first weight includes: Where P is the first weight of the growth point, exp() is an exponential function with the natural constant as the base, p is the frequency of the main direction of the ridge, norm{} is the normalization function, L is the number of grown points, and G ′ i G represents the grayscale feature of the second ridge at the i-th grown point. ′ i-1 For the grayscale feature of the second ridge at the (i-1)th grown point, k i Let k be the slope of the growth direction from the i-th grown point to the (i+1)-th grown point. i-1 It is the slope of the growth direction from the (i-1)th grown point to the ith grown point.

9. The method for intelligent analysis of geographic information from remote sensing and mapping images according to claim 1, characterized in that, The method for obtaining the grayscale features of the third ridge includes: The product of the adjusted weight and the second ridge grayscale feature is used as the third ridge grayscale feature.

10. The method for intelligent analysis of geographic information from remote sensing and mapping images according to claim 1, characterized in that, The step of determining whether the neighboring point is a new point to be grown based on the grayscale characteristics of the third ridge includes: If the grayscale feature of the third ridge is greater than the preset growth threshold, then the neighboring point is taken as a new growth point.

Images