A petroglyph recognition method and system
By acquiring the depth and location information of rock art inscriptions, marking valid inscriptions using inscription depth and location models, and establishing an information database for comparison, the problem of identification difficulties in traditional rock art identification has been solved, and efficient and accurate rock art information comparison has been achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 银川市贺兰山岩画管理处
- Filing Date
- 2026-02-05
- Publication Date
- 2026-06-05
AI Technical Summary
Traditional methods for detecting rock paintings rely on manual identification, which is affected by long-term wind and sand erosion, water erosion, and biological activity, leading to difficulties in identification and problems such as no search results or too many similar search results.
By acquiring the depth information of rock carvings, using the carving depth model to determine the positive and negative values of the carving depth information, marking valid and invalid carvings, and connecting valid carvings with the rock carving location model, a rock carving information database is established for comparison, and the rock carving number is output.
It simplifies the rock art identification process, improves the accuracy and efficiency of identification, reduces the difficulty of identification, and ensures accurate comparison of rock art information.
Smart Images
Figure CN122156567A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of rock art identification technology, and specifically relates to a rock art identification method and system. Background Technology
[0002] The Helan Mountain rock paintings are a precious cultural heritage left to us by our ancestors, possessing immense humanistic value. With a long history and preserved on exposed rock faces, they have long faced damage from wind, rain, salt erosion, and natural disasters, resulting in incomplete depictions. Traditional methods of examining the Helan Mountain rock paintings rely on manual inspection and photography, requiring highly specialized expertise from the personnel involved.
[0003] In the survey of rock paintings, after the initial discovery, a process involving cleaning, photographing, scanning, numbering, and archiving is required. If a survey or search for rock paintings occurs again after a considerable period, the existing rock paintings must be compared with the archived information to confirm their number and related information.
[0004] However, during each rock art survey or information gathering, professionals need to first determine the basic information of the rock art through graphic comparison and line proportions before searching the archived files. During this process, long-term wind and sand erosion, water dissolution, and biological activity make identification difficult for professionals, resulting in problems such as no results being found or too many similar search results. Summary of the Invention
[0005] Based on this, this application provides a rock art identification method and system to solve the problem that professional personnel have difficulty identifying rock art due to long-term wind and sand erosion, water erosion, biological activity and other factors, resulting in no search results or too many similar search results.
[0006] To achieve the above objectives, the present invention adopts the following solution: A method for identifying rock paintings, including...
[0007] Step S10: Obtain the depth information of the rock carving marks; Step S20: Based on the scratch depth model, determine the positive or negative value of the scratch depth information; If the value is positive, the depth information of the rock carving is marked as invalid. If the value is negative, the depth information of the rock carving is marked as a valid carving. Step S30: Obtain the location information of the rock carving marks; Step S40: Based on the rock painting inscription location model, preprocess the effective inscriptions and output the basic rock painting sample; Step S50: Based on the rock art information database, compare the similarity of basic rock art samples and output the rock art number.
[0008] Preferably, in step S20, the steps for establishing the scratch depth model are as follows: Step S21: Simulate the scratch depth, capture image information, and obtain simulated scratch parameters and image information parameters; Step S22: Input the simulated scratch parameters and image information parameters into the training model, and output the standard scratch depth information and the corresponding standard photo information; Step S23: Output the scratch depth model based on the standard scratch depth information and standard photograph information.
[0009] Preferably, step S40, establishing the location model of the rock carving marks, includes the following steps: Step S41: Obtain rock carving marks; Step S42: Based on the rock carving marks, input the location information of the rock carving marks corresponding to the rock carving marks; Step S43: Based on the rock carving marks and the corresponding rock carving mark location information, output the rock carving mark location model.
[0010] Preferably, the establishment of the rock art information database in step S50 includes the following steps: Step S51: Obtain the standard basic information of the rock paintings in the first general survey; Step S52: Establish a rock art information database based on standard basic information; The standard basic information of rock paintings includes the relative proportion of strokes and engravings in each rock painting and the standard location information of the corresponding rock painting engravings.
[0011] Preferably, in step S50, comparing the similarity of the basic rock painting samples includes the following steps: Step S53: Obtain the relative area ratio of the stroke marks of the rock painting basic sample; Step S54: Based on the rock art information database, compare the relative area ratio of the strokes and incised lines of the basic rock art samples and output the comparison list. Step S55: Sort the results based on the comparison list and output the first result.
[0012] Preferably, in step S50, comparing the similarity of the basic rock painting samples includes the following steps: Step S56: Based on the first result, obtain the relative area ratio of the stroke marks of the rock painting basic sample; Step S57: Input the scratch location information of the rock painting base sample, and compare whether the overlap between the scratch location information of the rock painting base sample and the standard scratch location information of the rock painting reaches the threshold. If so, output the rock painting number; If not, repeat steps S55 to S57, and proceed to the nth result based on the comparison list; where n = 1, 2, 3...n.
[0013] A rock art identification system, applied to any one of the rock art identification methods described above, comprising: The image acquisition module is used to acquire information on the depth and location of rock art inscriptions. The rock art information database module is used to store standard basic information on rock art from the first general survey; The notch depth model module is used to store different notch depths; The scratch depth judgment module, based on the scratch depth model module, judges the positive and negative values of the scratch depth information of the rock painting, and marks the judgment results as valid scratches and invalid scratches; The rock art inscription connection module, based on the location information of rock art inscriptions, is used to connect valid inscriptions with the location information of rock art inscriptions and output a basic sample of rock art. The rock art information comparison module, based on the rock art information database module, compares the rock art samples and outputs the rock art number.
[0014] Preferably, the rock art information comparison module includes: The relative proportion of stroke marks is calculated based on the rock art mark connection module to statistically analyze the relative area proportion of stroke marks in basic rock art samples. The proportion comparison unit, based on the rock art information database module, compares the relative area proportions of the strokes and inscriptions of the basic rock art samples and outputs a comparison list. The overlap comparison unit, based on the rock art information database module and the comparison list, compares the overlap of the location information of the basic rock art samples and outputs the rock art number. The technical solution adopted in this application can achieve the following beneficial effects: 1. By marking the valid and invalid marks on the damaged rock paintings, and then connecting the valid marks with the undamaged and accurate marks, the overall structure of the rock paintings becomes clearer. Then, by comparing it with information from previous surveys, the operation is simple and convenient, solving the problem that professional personnel have difficulty identifying the rock paintings due to long-term wind and sand erosion, water erosion, biological activity and other factors, and the problem of not being able to find the paintings or having too many similar search results.
[0015] 2. Simply put, the method used is perspective, which makes the damaged area transparent. By marking and connecting the damaged areas in the photo, influencing factors are reduced, the search difficulty is lowered, and thus the search efficiency and accuracy are improved. Attached Figure Description
[0016] Figure 1 This is a flowchart of the rock art identification method disclosed in the embodiments of this application.
[0017] Figure 2 This is a schematic diagram of the rock art recognition system disclosed in an embodiment of this application. Detailed Implementation
[0018] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings. Preferred embodiments of this application are shown in the drawings. However, this application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of this application.
[0019] It should be noted that when a device is considered to be "connected" to another device, it can be directly connected to the other device or there may be an intervening device present. The terms "inside," "top," "upper," "lower," "above," "below," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.
[0020] 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 application belongs. The terminology used herein in the specification of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0021] See Figure 1 This application provides a rock art identification method, including the following steps: Step S10, obtaining rock art engraving depth information; professionals take photos of the rock art, collect the photos, and obtain all engravings and their corresponding locations on the photos; focusing on the location of the damaged area, the following steps also mainly involve determining whether there are engravings and their locations at the damaged area.
[0022] Step S20: Based on the scratch depth model, determine the positive or negative value of the scratch depth information; by comparing the scratches on the acquired photo with the scratch depth model, determine whether the scratches on the photo are negative or positive.
[0023] If the value is positive, the depth information of the scratch is marked as invalid. If the value is positive, it means that the similar scratches in the photo are mostly raised rocks, scratches that have been eroded and smoothed, etc., which cause the scratches to appear raised or similar. Then, scratches with positive values are marked as invalid and are not considered.
[0024] If the value is negative, the depth information of the indentation is marked as a valid indentation, indicating that an indentation exists at the damaged area, and it is marked as a valid indentation. The plane with the largest relative proportion of the rock painting is taken as the zero point, with concave areas being negative values and convex areas being positive values.
[0025] In the above scheme, step S20, the steps for establishing the scratch depth model, are as follows: Step S21: Simulate scratch depth, capture image information, and obtain simulated scratch parameters and image information parameters; Step S22: Input the simulated scratch parameters and image information parameters into the training model, and output standard scratch depth information and corresponding standard photo information; Step S23: Based on the standard scratch depth information and standard photo information, output scratch depth model.
[0026] This method simulates scratches using a combination of mechanical and manual techniques, auxiliary tools, and acoustic distance measurement. By controlling the scratch depth, it obtains photographic information corresponding to different depths, i.e., simulated scratch parameters and image information parameters. All simulated scratch parameters and image information parameters are input into a neural network training model (preferably using acoustic distance measurement to obtain the specific scratch depth, or even 3D modeling based on acoustic radar). Before neural network training begins, the input data needs to be normalized to ensure that the distribution of training data and test data is consistent. Simultaneously, simulated damage occurs, and the scratch depth is detected and photographed using a professional scanning instrument. All parameters are then input into the neural network training model. Through multiple simulations, the simulated scratch depth, corresponding photographic information, simulated scratch damage parameters, and corresponding photographic information are normalized. Finally, the scratch information obtained by workers using professional equipment is input into the scratch depth model for comparison, and the comparison results are output. The neural network training model is a publicly available technology; the specific training process will not be detailed here.
[0027] Step S30: Obtain the location information of the rock carving marks; obtain the direction and specific location of the valid marks through the photograph.
[0028] Step S40: Based on the rock art inscription location model, preprocess the valid inscriptions to output the basic rock art sample; based on the photograph, obtain the inscription locations of the undamaged parts and the clear inscriptions close to the damaged parts to establish the rock art inscription location model; by using the direction and position of the inscriptions within the damaged parts, connect the nearest undamaged inscriptions using the nearest neighbor algorithm or curve fitting method to form a complete inscription; after all the inscriptions at the damaged parts have been connected, mark the inscriptions to form the basic rock art sample.
[0029] Step S50: Based on the rock art information database, compare the similarity of the basic rock art samples and output the rock art number; the rock art information database stores several rock art basic information that were discovered for the first time at least, left by professionals through methods such as photography, scanning, and copying, and has been numbered accordingly; compare and search the rock art basic sample formed this time with the rock art information database, and finally output the rock art number and corresponding basic information in the rock art information database corresponding to the rock art basic sample formed this time.
[0030] Furthermore, a rock art database was first established, summarizing information from the first or subsequent surveys. This database includes several rock paintings, their distribution locations, the location, direction, and area percentage of each carving. Over extended periods, the rock paintings in the field are weathered and relocated by mudslides, wind, etc. When professionals locate a rock painting, it is found to be weathered, covered by vegetation, and silt deposited. After simple processing, the professionals photograph it. Through photo processing, the carving marks in the damaged areas are extracted. Then, by comparing with the scratch depth model, the positive and negative values of the scratch depth information in the damaged area are determined, thereby identifying the valid scratches. After all the valid scratches in the damaged areas have been marked, the valid scratches and the scratches in the photos that have not been damaged by vegetation are connected and marked to obtain the basic sample of the rock art. After obtaining the basic sample of the rock art, it is compared with the rock art in the rock art information database by means of scratch proportion, rock art image, scratch overlap, etc., so as to output that the rock art surveyed by professionals belongs to the rock art with the corresponding number in the rock art information database.
[0031] The technical solution of the rock art identification method adopted in this application can achieve the following beneficial effects: 1. By marking the valid and invalid marks on the damaged rock paintings, and then connecting the valid marks with the undamaged and accurate marks, the overall structure of the rock paintings becomes clearer. Then, by comparing it with information from previous surveys, the operation is simple and convenient, solving the problem that professional personnel have difficulty identifying the rock paintings due to long-term wind and sand erosion, water erosion, biological activity and other factors, and the problem of not being able to find the paintings or having too many similar search results.
[0032] 2. Simply put, the method used is perspective, which makes the damaged area transparent. By marking and connecting the damaged areas in the photo, influencing factors are reduced, the search difficulty is lowered, and thus the search efficiency and accuracy are improved.
[0033] In the above scheme, step S40, the establishment of the rock carving location model, includes the following steps: Step S41: Obtain rock carving marks; Step S42: Based on the rock carving marks, input the location information of the corresponding rock carving marks; Step S43: Based on the rock carving marks and the corresponding location information of the rock carving marks, output the location model of the rock carving marks.
[0034] The engravings on undisturbed rock art are recorded using coordinates. Coordinates are established based on the rock art's prominent location, center, or corners. Lines are drawn connecting the start and end points of each engraving, along with several points along its path, to form the engraving itself. A coordinate system is established on photographs, and each engraving is marked to determine its coordinate information. When professionals confirm the valid engravings in the damaged areas of the rock art discovered in this case, they are also represented as coordinate points. These coordinates are then connected to the engravings in the undisturbed areas using smooth curves to form a complete rock art structure. This method is simple and more precise.
[0035] In another embodiment of this application, step S51 involves obtaining the standard basic information of the rock paintings in the first general survey; step S52 involves establishing a rock painting information database based on the standard basic information; wherein, the standard basic information of the rock paintings includes the relative proportion of strokes and engravings in each rock painting and the corresponding standard position information of the engravings.
[0036] After discovering several rock paintings through investigation, professionals used instruments to scan, photograph, and scan the engravings to record and number each rock painting, and then compiled them. The compiled information was stored and a rock painting information database was established. The rock painting information database is based on parameters such as the number of each rock painting, the type of engraving, the location of the engraving, the depth of the engraving, and the relative proportion of the engraving area.
[0037] Further, in step S50, comparing the similarity of the rock painting base samples includes the following steps: step S53, obtaining the relative area ratio of the strokes and incised lines of the rock painting base samples; step S54, based on the rock painting information database, comparing the relative area ratio of the strokes and incised lines of the rock painting base samples and outputting a comparison list; step S55, sorting based on the comparison list and outputting the first result.
[0038] After simple cleaning by professionals, images are captured using specialized equipment and presented as photographs. The types of engravings and their corresponding relative proportions are obtained and then compared with the types of engravings and their corresponding relative proportions for each rock painting in the rock painting database to identify several similar or nearly identical rock paintings. The above steps are repeated, and a unit area is randomly selected from another photograph for comparison and sorting. Through multiple comparisons, the ten closest rock paintings are finally sorted and a comparison list is output. The first result on the comparison list is the preferred result.
[0039] Furthermore, in step S50, comparing the similarity of the rock painting base samples includes the following steps: Step S56, based on the first result, obtain the relative area ratio of the strokes and inscriptions of the rock painting base sample; Step S57, input the inscription location information of the rock painting base sample, and compare whether the overlap between the inscription location information of the rock painting base sample and the standard location information of the rock painting inscription reaches a threshold; if yes, output the rock painting number; if no, repeat steps S55 to S57, and based on the comparison list, continue to the nth result; where n=1, 2, 3...n.
[0040] Using the output comparison list, select the first result from the list, then select the type of incised lines and their corresponding proportions on the photograph of the rock painting, and determine the location of the incised lines. At the same time, mark the type of incised lines and their corresponding proportions at the corresponding locations on the rock painting image of the first result using the same method (e.g., coordinate method) as the incised line locations mentioned above. Then compare the overlap of the incised line locations. Similarly, repeat this process multiple times, selecting different locations and comparing unit areas to output the comparison results. If the overlap of the comparison results is low, much less than the preset threshold, then compare the rock painting image information obtained by the professional this time with the second result on the comparison list, until the comparison results stop within the preset threshold, and output the number and other relevant information of the nth result on the comparison list.
[0041] The rock art identification method of this application determines valid engravings through engraving depth information. Then, based on the engraving direction of the valid engravings, the engravings at the undamaged locations of the rock art are connected to form a relatively complete rock art structure. Next, by determining the location and the engraving type and proportion at each location, the engraving type and proportion at each location are compared with the corresponding engraving type and proportion at each location in all rock art databases, ultimately forming a comparison list (the closer the proportion, the higher the ranking). Then, the engraving type and corresponding proportion of the image acquired by professionals are compared with the engraving type and corresponding proportion of the engraving location information according to the arrangement order of the comparison list, thereby determining the degree of overlap of the engraving locations. When the degree of overlap of the engraving locations reaches a preset threshold, the rock art number and related information in the rock art database are output.
[0042] Please see Figure 2This application also provides a rock art recognition system, applied to any of the rock art recognition methods described above, comprising: an image acquisition module for acquiring rock art indentation depth information and rock art indentation location information; a rock art information database module for storing standard basic information of rock art from the first general survey; an indentation depth model module for storing different indentation depths and their corresponding standard photograph information; an indentation depth judgment module for judging the positive and negative values of rock art indentation depth information based on the indentation depth model module, and marking the judgment results as valid and invalid indentations; a rock art indentation connection module for connecting valid indentations with rock art indentation location information, and outputting a basic rock art sample; and a rock art information comparison module for comparing the basic rock art sample based on the rock art information database module and outputting a rock art number.
[0043] For example, if a rock painting is in the shape of a five-pointed star, with one point damaged, the valid engraving (one point of the five-pointed star) is identified and connected to the undamaged part of the five-pointed star to form a complete five-pointed star.
[0044] The output of the image acquisition module is connected to the input of the rock art information comparison module and the input of the scratch judgment module, and is used to transmit the acquired rock art scratch depth information and rock art scratch location information to the rock art information comparison module.
[0045] The output of the rock art information database module is connected to the input of the scratch judgment module, and is used to store the standard basic information of rock art from the first or subsequent surveys. The rock art information comparison module can retrieve scratch depth information, rock art scale position information, etc. from the rock art information database module.
[0046] The output of the scratch depth model module is connected to the input of the scratch judgment module, and is used to learn and simulate scratch depth through the neural network training system; the rock painting information comparison module can retrieve the standard scratch depth information and standard photo information output after training.
[0047] The input of the scratch judgment module is connected to the output of the image acquisition module, the output of the rock art information database module, and the output of the scratch depth model module. The output of the scratch judgment module is connected to the input of the rock art scratch connection module, which is used to connect valid scratches with undamaged scratches.
[0048] The output end of the rock art inscription connection module is connected to the input end of the rock art information comparison module. The rock art information comparison module includes: a stroke inscription relative proportion unit, a proportion comparison unit, and an overlap comparison unit. The input end of the stroke inscription relative proportion unit is connected to the output end of the rock art information database module and the output end of the rock art inscription connection module, respectively. Based on the rock art inscription connection module, the relative area proportion of stroke inscriptions in the basic rock art samples is statistically analyzed.
[0049] The input of the proportion comparison unit is connected to the output of the stroke inscription relative proportion unit. It compares the relative area proportion of the stroke inscriptions of the rock painting base sample and outputs a comparison list.
[0050] The input of the overlap comparison unit is connected to the output of the proportion comparison unit and the output of the rock art information database module. Based on the comparison list output by the rock art information database module and the proportion comparison unit, the overlap of the strokes of the basic rock art samples in relative positions is compared, and the rock art number is output.
[0051] Furthermore, the degree of overlap is determined using methods such as facial recognition, object recognition by photo, and photo scanning comparison. These methods are all existing technologies and will not be described in detail here. This application preferably uses facial recognition to determine the degree of overlap.
[0052] The above-described embodiments merely illustrate the device deployment method of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the patent application. It should be noted that for those skilled in the art, several adjustments and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. A method for identifying rock paintings, characterized in that, Includes the following steps: Step S10: Obtain the depth information of the rock carving marks; Step S20: Based on the scratch depth model, determine the positive or negative value of the scratch depth information; If the value is positive, the depth information of the rock carving is marked as invalid. If the value is negative, the depth information of the rock carving is marked as a valid carving. Step S30: Obtain the location information of the rock carving marks; Step S40: Based on the rock painting inscription location model, preprocess the effective inscriptions and output the basic rock painting sample; Step S50: Based on the rock art information database, compare the similarity of basic rock art samples and output the rock art number.
2. The rock art identification method according to claim 1, characterized in that, In step S20, the steps for establishing the scratch depth model are as follows: Step S21: Simulate the scratch depth, capture image information, and obtain simulated scratch parameters and image information parameters; Step S22: Input the simulated scratch parameters and image information parameters into the training model, and output the standard scratch depth information and the corresponding standard photo information; Step S23: Output the scratch depth model based on the standard scratch depth information and standard photograph information.
3. The rock art identification method according to claim 1, characterized in that, Step S40, the establishment of the rock carving location model includes the following steps: Step S41: Obtain rock carving marks; Step S42: Based on the rock carving marks, input the location information of the rock carving marks corresponding to the rock carving marks; Step S43: Based on the rock carving marks and the corresponding rock carving mark location information, output the rock carving mark location model.
4. The rock art identification method according to claim 1, characterized in that, The establishment of the rock art information database in step S50 includes the following steps: Step S51: Obtain the standard basic information of the rock paintings in the first general survey; Step S52: Establish a rock art information database based on standard basic information; The standard basic information of rock paintings includes the relative proportion of strokes and engravings in each rock painting and the standard location information of the corresponding rock painting engravings.
5. The rock art identification method according to claim 4, characterized in that, In step S50, comparing the similarity of the basic rock painting samples includes the following steps: Step S53: Obtain the relative area ratio of the stroke marks of the rock painting basic sample; Step S54: Based on the rock art information database, compare the relative area ratio of the strokes and incised lines of the basic rock art samples and output the comparison list. Step S55: Sort the results based on the comparison list and output the first result.
6. The rock art identification method according to claim 5, characterized in that, In step S50, comparing the similarity of the basic rock painting samples includes the following steps: Step S56: Based on the first result, obtain the relative area ratio of the stroke marks of the rock painting basic sample; Step S57: Input the scratch location information of the rock painting base sample, and compare whether the overlap between the scratch location information of the rock painting base sample and the standard scratch location information of the rock painting reaches the threshold. If so, output the rock painting number; If not, repeat steps S55 to S57, and proceed to the nth result based on the comparison list; where n = 1, 2, 3...n.
7. A rock art identification system, characterized in that, The rock art identification method applied to any one of claims 1 to 6 includes: The image acquisition module is used to acquire information on the depth and location of rock art inscriptions. The rock art information database module is used to store standard basic information on rock art from the first general survey; The notch depth model module is used to store different notch depths; The scratch depth judgment module, based on the scratch depth model module, judges the positive and negative values of the scratch depth information of the rock painting, and marks the judgment results as valid scratches and invalid scratches; The rock art inscription connection module, based on the location information of rock art inscriptions, is used to connect valid inscriptions with the location information of rock art inscriptions and output a basic sample of rock art. The rock art information comparison module, based on the rock art information database module, compares the rock art samples and outputs the rock art number.
8. The rock art identification system according to claim 7, characterized in that, The rock art information comparison module includes: The relative proportion of stroke marks is calculated based on the rock art mark connection module to statistically analyze the relative area proportion of stroke marks in basic rock art samples. The proportion comparison unit, based on the rock art information database module, compares the relative area proportions of the strokes and inscriptions of the basic rock art samples and outputs a comparison list. The overlap comparison unit, based on the rock art information database module and the comparison list, compares the overlap of the location information of the basic rock art samples and outputs the rock art number.