Method and system for intelligent verification of land change evidence photos based on multi-source information fusion

Abstract

The application discloses a method for intelligent verification of land change evidence photos based on multi-source information fusion, and has the characteristics that a to-be-verified image set, a travel database and a land type database are acquired respectively; travel features are determined according to the to-be-verified image set and the travel database; whether the to-be-verified image set is abnormal is judged according to the travel features and the travel database; when the to-be-verified image set is abnormal, the to-be-verified image set is marked as wrong; when the to-be-verified image set is not abnormal, first land feature types in a to-be-verified area are determined according to the land type database; second land feature types in each photo in the to-be-verified image set are determined according to the photos; and qualified to-be-verified image sets are determined according to the first land feature types in the to-be-verified area and the second land feature types in each photo in the to-be-verified image set. The application constructs a multi-level automatic verification framework, reduces the problem of false evidence from the technical level, and effectively improves the verification efficiency of evidence photos.

Description

Technical Field

[0001] This invention relates to the field of land survey technology, and in particular to a method and system for intelligent verification of land change evidence photos based on multi-source information fusion. Background Technology

[0002] The annual land use change survey is a special project conducted by China's Ministry of Natural Resources to understand annual changes in land use. Its aim is to update the land database and create a unified base map for land spatial management. The survey acquires land cover data using satellite and aerial remote sensing technologies, and combines this data with county-level on-site verification and provincial-level checks to form a dynamic record. The survey results provide basic data support for land spatial planning assessments, farmland protection evaluations, and other related tasks.

[0003] In the field verification work of land change surveys, the evidence photos taken on site are the core basis for determining the land type, boundaries and authenticity of changes of land parcels.

[0004] Currently, national-level internal verification relies entirely on manual interpretation of massive amounts of photographs by auditors. Due to workload and limited personnel, it is difficult to effectively verify whether photographs were taken on-site in real time, posing a risk of using historical or out-of-town photographs for false evidence, resulting in fundamental loopholes in the verification process. The verification process is highly dependent on staff experience; staff manually judge whether the features in the photographs match the "claimed land type / scope" recorded in the database. This is influenced by individual staff experience and focus, making it difficult to standardize standards and prone to oversights and misjudgments. With millions of photographs being used as evidence, a purely manual verification model is time-consuming and labor-intensive. The limited number of staff and the large workload in the current verification model lead to low efficiency in verifying evidence photographs. Summary of the Invention

[0005] The main objective of this invention is to provide an intelligent verification method and system for land use change evidence photos based on multi-source information fusion, aiming to solve the problem of low efficiency in evidence photo verification.

[0006] To achieve the above objectives, the present invention provides an intelligent verification method for land use change evidence photos based on multi-source information fusion, comprising: Obtain the image set to be verified, the travel database, and the land category database respectively; Based on the set of images to be verified and the travel database, travel characteristics are determined; Based on the trip characteristics and the trip database, determine whether the image set to be checked contains any anomalies: When the set of images to be verified contains an anomaly, the set of images to be verified will be marked as incorrect; When there are no anomalies in the image set to be verified, the first land cover type in the area to be verified is determined according to the land cover database; Based on each of the photographs in the image set to be verified, determine the type of each second land feature within the photograph; Based on the first land cover type in the area to be verified and the second land cover type in each of the photos in the image set to be verified, the qualified image set to be verified is determined.

[0007] Preferably, the step of determining the travel features based on the set of images to be verified includes: Based on the image set to be verified, determine the first coordinate information and the absolute timestamp of each photo in the image set to be verified. The path order is determined based on the first coordinate information and the trip database; Based on the path order and each absolute timestamp, determine the time interval between two adjacent absolute timestamps in each absolute timestamp; Determine the time proportion based on each of the aforementioned time intervals; Based on the path order and each of the first coordinate information, determine the interval distance between two adjacent first coordinate information in each of the first coordinate information; Determine the distance ratio based on each of the aforementioned intervals; Determine whether the time ratio and the distance ratio are consistent; When the time ratio and the distance ratio are inconsistent, the image set to be checked will be labeled as incorrect. When the time ratio and the distance ratio are consistent, each of the interval time periods is used as the travel feature.

[0008] Preferably, the step of determining whether the image set to be checked has any anomalies based on the trip features and the trip database includes: Based on the first coordinate information, determine the region information of the image set to be verified; Based on the regional information and the first coordinate information, determine the traffic attributes of each travel feature; Based on the aforementioned travel database and regional information, several movement characteristics are obtained; Determine whether the traffic attributes and movement speed of the travel characteristics simultaneously meet one of the movement characteristics; When both the traffic attribute and movement speed of the travel characteristics meet any one of the movement characteristics, the following step is executed: If there are no anomalies in the image set to be verified, the step of determining the first land cover type in the area to be verified based on the land cover database is executed: When neither the traffic attribute nor the speed of movement of the trip features matches any of the movement features, the step of labeling the image set to be checked as incorrect is executed.

[0009] Preferably, the step of labeling the image set to be verified as incorrect when the time ratio and the distance ratio are inconsistent includes: When the time ratio and the distance ratio are inconsistent, determine the ratio result; Based on the similarity of the ratio results, several ratio groups are determined; Determine whether the number of each ratio group is greater than a first preset number; When the number of each ratio group is greater than the first preset number, the step of labeling the image set to be checked as incorrect is executed. When the number of each ratio group is less than or equal to the first preset number, several groups are determined according to each ratio group and each time interval. Each of the aforementioned groups is respectively identified as the travel feature.

[0010] Preferably, the step of determining whether the image set to be checked has any anomalies based on the trip features and the trip database includes: Based on the aforementioned travel database, several movement features are obtained; Determine whether each of the aforementioned travel features contains a matching movement feature; When all the aforementioned travel features match the specified movement features, the following step is executed: if the image set to be verified does not contain any anomalies, the step of determining the first land cover type in the area to be verified based on the land cover database is performed: If any of the described travel features does not match the described movement feature, the step of labeling the image set to be checked as incorrect is executed.

[0011] Preferably, the step of determining the type of each second land feature within each photograph in the image set to be verified includes: Based on the set of images to be verified, determine the pixel blocks in each of the photos in the set of images to be verified; Obtain confirmation information, and determine the type of each second land feature in the photo based on the confirmation information and each pixel block; Preferably, the step of determining a qualified image set to be verified based on each first land feature type in the area to be verified and each second land feature type in each photograph of the image set to be verified includes: Based on each first land feature type in the area to be verified and each second land feature type in each photograph in the image set to be verified, determine whether the number of similarities between each second land feature type and each first land feature type is greater than a second preset number; When the number of similarities between each of the second land cover types and each of the first land cover types is greater than or equal to the second preset number, the step of determining the qualified set of images to be checked is executed; When the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number, the image set to be verified is marked as pending verification.

[0012] Preferably, the step of marking the image set to be verified as pending verification when the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number includes: When the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number, the photo is marked as abnormal; Determine whether the ratio of the photos marked as abnormal to the sum of the photos in the image set to be checked is greater than a preset ratio; When the ratio of the photos marked as abnormal to the sum of the photos in the image set to be verified is greater than or equal to the preset ratio, the step of marking the image set to be verified as pending verification is executed. When the ratio of the photos marked as abnormal to the sum of the photos in the image set to be verified is less than the preset ratio, the step of determining the qualified image set to be verified is performed.

[0013] Preferably, the step of determining the path order based on each of the first coordinate information and the trip database includes: Based on the land category database, determine the coordinate region within the area to be verified; Determine whether each of the first coordinate information is located within the coordinate region; When all the first coordinate information is located within the coordinate area, the path order is determined based on each of the first coordinate information and the trip database; When any of the first coordinate information is outside the coordinate region, the step of labeling the image set to be checked as incorrect is executed.

[0014] To achieve the above objectives, based on the intelligent verification system for land change evidence photos based on multi-source information fusion, the intelligent verification method for land change evidence photos based on multi-source information fusion as described above is applied, including an execution module, a judgment module, and a calculation module; The execution module is used to acquire the image set to be verified, the travel database, and the land category database respectively; and to determine the travel characteristics based on the image set to be verified and the travel database. The judgment module is used to: determine whether the image set to be verified has any abnormalities based on the travel characteristics and the travel database; when the image set to be verified has any abnormalities, mark the image set to be verified as incorrect; when the image set to be verified does not have any abnormalities, determine the type of each first land feature in the area to be verified based on the land type database. The calculation module is used to determine the types of second land features in each of the photos in the image set to be verified; and to determine the qualified image set to be verified based on the types of first land features in the area to be verified and the types of second land features in each of the photos in the image set to be verified.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: By deeply integrating and analyzing the travel characteristics and land feature types of each photo in the image set to be verified, a multi-level automated verification framework was constructed. This minimizes the problem of false evidence from a technical perspective and effectively improves the efficiency of the verification of evidence photos. Attached Figure Description

[0016] To more clearly illustrate the technical solutions 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 the structures shown in these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an embodiment of the intelligent verification method for land use change evidence photos based on multi-source information fusion according to the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0020] The following describes, with reference to the accompanying drawings, an intelligent verification method and system for land use change evidence photos based on multi-source information fusion, according to embodiments of the present invention.

[0021] Figure 1 This is a flowchart illustrating an embodiment of the intelligent verification method for land use change evidence photos based on multi-source information fusion according to the present invention.

[0022] Please see Figure 1 To achieve the above objectives, the intelligent verification method for land use change evidence photos based on multi-source information fusion provided in the first embodiment of the present invention includes: Step S10: Obtain the image set to be verified, the travel database, and the land category database, respectively; Step S20: Determine the travel characteristics based on the image set to be verified and the travel database; Step S30: Based on the trip characteristics and the trip database, determine whether there are any anomalies in the image set to be checked. Step S40: When there is an anomaly in the image set to be checked, the image set to be checked is marked as incorrect; Step S50: When there are no anomalies in the image set to be verified, determine the type of each first land feature in the area to be verified based on the land type database; Step S60: Determine the types of second land features in each photo based on the photos in the image set to be verified; Step S70: Determine the qualified image set to be verified based on the first land cover type in the area to be verified and the second land cover type in each photo in the image set to be verified.

[0023] By deeply integrating and analyzing the travel characteristics and land feature types of each photo in the image set to be verified, a multi-level automated verification framework was constructed. This minimizes the problem of false evidence from a technical perspective and effectively improves the efficiency of the verification of evidence photos.

[0024] Specifically, the travel database is set according to the region and includes at least one of the following: map information of the image set to be verified, walking distance per hour, driving distance per hour, and mountain climbing distance per hour. Different modes of movement are categorized into different movement characteristics, providing a standard basis for the time taken by staff to travel routes. Each movement characteristic is an interval speed.

[0025] In the second embodiment of the present invention, based on the first embodiment, step S20 includes: Step S21: Based on the image set to be verified, determine the first coordinate information and the absolute timestamp of each photo in the image set to be verified. Step S22: Determine the path sequence based on the first coordinate information and the travel database; Step S23: Determine the time interval between two adjacent absolute timestamps based on the path order and each absolute timestamp. Step S24: Determine the time proportion based on each time interval. Step S26: Determine the interval distance between two adjacent first coordinate information in each first coordinate information according to the path order and each first coordinate information; Step S27: Determine the distance ratio based on each interval distance; Step S28: Determine whether the time ratio and distance ratio are consistent; Step S29: When the time scale and distance scale are inconsistent, the image set to be checked is labeled incorrectly. Step S210: When the time ratio and distance ratio are consistent, each time interval is used as a travel feature.

[0026] Based on the path order, the first coordinate information of each photo, and the absolute timestamp of each photo, the time interval between two adjacent absolute timestamps is determined, thereby determining the time ratio; then the distance ratio is determined in turn. Based on the distance ratio and the time ratio, the authenticity of the photo is quickly determined. Based on the first coordinate information and the absolute timestamp, false or incorrect photos are intelligently filtered out. For example, if the journey takes 30 minutes, but the time difference between the photos is only 5 minutes, it is judged as a spatiotemporal contradiction, and the image set to be verified is incorrectly labeled.

[0027] In the third embodiment of the present invention, based on the second embodiment, step S30 includes: Step S31: Determine the region information of the image set to be checked based on the first coordinate information; Step S32: Determine the traffic attributes of each travel feature based on the regional information and each first coordinate information; Step S33: Obtain several movement features based on the travel database and regional information; Step S33: Determine whether the traffic attribute and speed of the trip feature simultaneously meet one of the travel features; Step S34: When both the traffic attribute and movement speed of the travel characteristics meet any of the movement characteristics, proceed to the step of determining the first land cover type in the area to be verified based on the land cover database when there are no anomalies in the image set to be verified. Step S35: When neither the traffic attribute nor the speed of the travel feature matches any of the travel features, proceed with the step of labeling the image set to be checked incorrectly.

[0028] Traffic attributes are determined based on map information of the image set to be verified. For example, when the route corresponding to the travel feature can only climb mountains, the traffic attribute of the travel feature is only mountain climbing; when the route corresponding to the travel feature has highways, the traffic data of the travel feature simultaneously includes multiple modes of transportation such as cars, walking and electric vehicles.

[0029] The process involves judging the travel features based on various movement characteristics within the travel database. Only when both the traffic and movement attributes of the travel feature match the movement characteristics can the correctness of the image set under verification be proven, thus proceeding to the next step. This step further filters out abnormal data. For example, the movement characteristics of walking are 3 km / h to 5 km / h, and the movement characteristics of mountain climbing are 1.5 km / h to 2 km / h. When the traffic attribute of the travel feature is mountain climbing with a movement speed of 10 km / h, but the movement characteristics of mountain climbing are 1.5 km / h to 2 km / h, the recognition fails, proving that this travel feature is abnormal data and needs to be labeled incorrectly. Similarly, another travel feature is walking and mountain climbing with a movement speed of 4 km / h. The traffic data for this travel feature includes walking, and the movement speed is within the walking movement characteristic range, so the recognition passes and proceeds to the next step.

[0030] In the fourth embodiment of the present invention, based on the second embodiment, step S29 includes: Step S211: When the time ratio and distance ratio are inconsistent, determine the ratio result; Step S212: Determine several ratio groups based on the similarity of the ratio results; Step S213: Determine whether the number of each ratio group is greater than the first preset number; Step S214: When the number of each ratio group is greater than the first preset number, execute the step of labeling the image set to be checked with errors. Step S215: When the number of each ratio group is less than or equal to the first preset number, determine several groups according to each ratio group and each time interval. Step S216: Each group is identified as a travel feature.

[0031] Since the entire photo-taking process may involve a mix of cars and pedestrians, identifying multiple groups as travel characteristics can further reduce the probability of false positives. Generally, the entire photo-based evidence collection process will not exceed three groups; therefore, if the number of groups is too large, the image set to be verified can be identified as a fake image set and marked as incorrect.

[0032] In the fifth embodiment of the present invention, based on the fourth embodiment, step S30 includes: Step S35: Obtain several movement features based on the travel database; Step S36: Determine whether all travel features have the same movement feature; Step S37: When all travel features have the same movement feature, proceed to the step of determining the first land cover type in the area to be verified based on the land cover database when the image set to be verified has no anomalies. Step S38: When no travel feature has the same movement feature, perform the step of labeling the image set to be checked incorrectly.

[0033] Multiple travel features are evaluated separately. Only when all travel features exhibit matching movement characteristics can the correctness of the image set under verification be proven. This prevents the inclusion of fake photos in the image set and ensures the maximum possible authenticity of the entire image set, thus proceeding to the next step, which further filters out abnormal data. A single travel feature must contain at least one group. Only when all groups within a single travel feature exhibit the same movement characteristics can that travel feature proceed to the next step. For example, if a travel feature includes two groups, one with a movement speed of 2 km / h and the other with a movement speed of 5 km / h, and one group is identified as mountain climbing and the other as walking, this travel feature will proceed to the next step upon successful identification.

[0034] In the sixth embodiment of the present invention, based on any one of the second to fifth embodiments, step S60 includes: Step S61: Based on the image set to be verified, determine the pixel blocks in each photo in the image set to be verified; Step S62: Obtain confirmation information, and determine the types of second land features in the photo based on the confirmation information and each pixel block.

[0035] Specifically, step S62 includes: Step S63: Obtain confirmation information within a third preset number of photos in the current image set to be verified; and determine the second land feature type within the third preset number of photos based on the confirmation information within each of the third preset number of photos and the pixel blocks within each of the third preset number of photos in the current image set to be verified. Step S64: Determine the type of second land cover in each pixel block of the current image set to be checked based on the type of second land cover in each photo in the third preset quantity and each pixel block in the current image set to be checked. Step S65: Determine the second land cover type of the remaining photos in the current image set to be verified based on the second land cover type of each pixel block in the current image set to be verified and the remaining photos in the current image set to be verified. Step S66: Determine the type of second land cover in each photo in the current image set to be verified based on the type of second land cover in the third preset number of photos in the current image set to be verified and the type of second land cover in the remaining photos in the current image set to be verified.

[0036] The image set to be verified covers a relatively small area (similar to a 20km or 50km area), with generally similar terrain and landforms. By manually annotating and confirming information from a certain number of photos (i.e., the third preset number of photos) extracted from the entire image set to be verified, and aligning the pixel blocks and land cover types in the image set to be verified, the remaining photos are automatically standardized, enabling rapid annotation of the remaining photos in the image set to be verified.

[0037] Specifically, after step S50, the following steps are included: Step S51: Determine whether the interval between the first coordinate information in each photo is less than the preset interval; Step S52: When the interval between the first coordinate information in at least two photos is less than a preset interval, the photos with an interval less than the preset interval are combined into a combined image, the remaining photos in the combined image are excluded from the images to be checked, and all combined images are treated as photos and step S61 is executed.

[0038] Step S53: When the interval distance of the first coordinate information in each photo is greater than or equal to the preset interval, proceed to step S61.

[0039] A composite image is a cluster of photos stitched together with intervals less than a preset interval. By grouping the composite images together with the remaining photos using first coordinate information intervals less than a preset interval, the second land cover type is determined, ensuring that multiple photos within the same first coordinate information pass the evidentiary verification. For example, if a composite image includes five photos, one of which contains a road and the remaining photos do not, since it is a composite image, the road can be considered the second land cover type in the output.

[0040] Specifically, step S52 includes: Step S54: When the interval between the first coordinate information in at least two photos is less than a preset interval, combine the photos with an interval less than the preset interval into a combined image. Step S55: Based on the land use database, determine the location of the first land feature of each first land feature type in the area to be verified; Step S56: Determine the location of the second land feature based on the positional relationship of each second land feature type in the combined photograph; Step S57: Determine the location of the third feature based on the positional relationship of each second feature type in the remaining photos after excluding the combined images of the images to be checked. Step S58: Determine whether the similarity rate of the second or third feature location relative to the first feature location is greater than the preset similarity rate. Step S59: When the similarity rate of the second feature location and / or the third feature location relative to the first feature location is greater than or equal to the preset similarity rate, execute step S53 on the photo corresponding to the second feature location and / or the photo corresponding to the third feature location. Step S510: When the similarity rate between the second feature location and / or the third feature location and the first feature location is less than the preset similarity rate, mark the photo corresponding to the second feature location and / or the photo corresponding to the third feature location as pending verification.

[0041] The location of the second type of land cover in the photo is re-verified to ensure the accuracy of the final verification score. For example, if the locations of the first land cover are A, B, C, D, and F in sequence, but the location of the second land cover is B, C, D, F, and A, the similarity rate is high and the judgment is passed. However, if the location of the third land cover is C, A, F, B, and D, the similarity rate is low and the judgment is failed and marked as pending verification.

[0042] In the seventh embodiment of the present invention, based on the sixth embodiment, step S70 includes: Step S71: Based on the first land cover type in the area to be verified and the second land cover type in each photo in the image set to be verified, determine whether the similarity between each second land cover type and each first land cover type is greater than the second preset number. Step S72: When the number of similarities between each second land cover type and each first land cover type is greater than or equal to the second preset number, the step of determining a qualified set of images to be verified is executed. Step S73: When the number of similarities between each second land cover type and each first land cover type is less than the second preset number, the image set to be verified is marked as pending verification.

[0043] Specifically, the land use database includes all land use types, the coordinate regions of the area to be verified, and the locational relationships of various land cover types within the area to be verified.

[0044] The pixel blocks in the photo are marked (similar to circling similar pixels in an image, with the marked pixel blocks inside the boxes). The types of pixel blocks are then manually selected to determine the types of second land features in the photo. The types of first land features in the area to be verified are then compared with those in the land use database. Only when the number of similarities between the types of second land features in the photo and the types of first land features in the area to be verified reaches a second preset number can the image set to be verified be marked as qualified.

[0045] The image set to be verified for land use classification has already been verified for authenticity through the timeline. This stage mainly verifies whether the photos can be used. Therefore, the images marked as pending verification can be provided again by the staff as qualified images.

[0046] In the eighth embodiment of the present invention, based on the seventh embodiment, step S73 includes: Step S74: When the number of similarities between each second land cover type and each first land cover type is less than the second preset number, the photo is marked as abnormal. Step S75: Determine whether the ratio of each photo marked as abnormal to the sum of each photo in the image set to be checked is greater than a preset ratio. Step S76: When the ratio of each photo marked as abnormal to the sum of each photo in the image set to be verified is greater than or equal to a preset ratio, the step of marking the image set to be verified as pending verification is executed. Step S77: When the ratio of the photos marked as abnormal to the sum of the photos in the image set to be checked is less than a preset ratio, the step of determining the qualified image set to be checked is executed.

[0047] All photos were checked again, and those that were authentic but not of good quality were screened out to ensure the authenticity and accuracy of the final verification.

[0048] By repeatedly verifying data over time and mileage, and using technological means to eliminate false and invalid evidence at the source, the reliability of land use change survey results is greatly improved. Automated verification forms a routine technical supervision capability, creating a strong deterrent against potential violations and purifying the survey operation environment.

[0049] The sets of images marked as pending verification and those marked as incorrect are returned. Staff only need to manually check the returned and revised photos before resubmission. This allows staff to focus their manual review efforts on high-risk photos, greatly reducing the difficulty and intensity of manual work.

[0050] In the ninth embodiment of the present invention, based on the seventh embodiment, step S22 includes: Step S217: Determine the coordinate region of the area to be verified based on the land category database; Step S218: Determine whether each first coordinate information is located within the coordinate region; Step S219: When all the first coordinate information is within the coordinate area, determine the path order based on the first coordinate information and the travel database; Step S220: When any first coordinate information is outside the coordinate region, execute the step of labeling the image set to be checked with errors.

[0051] By verifying the first coordinate information of each image, it is ensured that the first coordinate information when the photo was taken is within the coordinate area of ​​the area to be verified, thereby further reducing false evidence submission.

[0052] To achieve the above objectives, based on the intelligent verification system for land change evidence photos based on multi-source information fusion, the intelligent verification method for land change evidence photos based on multi-source information fusion as described above is applied, including an execution module, a judgment module, and a calculation module; The execution module is used to acquire the image set to be verified, the travel database, and the land category database respectively; and to determine the travel characteristics based on the image set to be verified and the travel database. The judgment module is used to: determine whether the image set to be verified has any abnormalities based on the travel characteristics and the travel database; when the image set to be verified has any abnormalities, mark the image set to be verified as incorrect; when the image set to be verified does not have any abnormalities, determine the type of each first land feature in the area to be verified based on the land type database. The calculation module is used to determine the types of second land features in each of the photos in the image set to be verified; and to determine the qualified image set to be verified based on the types of first land features in the area to be verified and the types of second land features in each of the photos in the image set to be verified.

[0053] In the description of this specification, references to terms such as "one embodiment," "another embodiment," "other embodiments," or "first embodiment to Xth embodiment," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example that are included in at least one embodiment or example of the present invention.

[0054] In this specification, the illustrative expressions of the terms used above do not necessarily refer to the same embodiments or examples.

[0055] Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0056] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0057] The sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0058] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for intelligent verification of land use change evidence photos based on multi-source information fusion, characterized in that: include: Obtain the image set to be verified, the travel database, and the land category database respectively; Based on the set of images to be verified and the travel database, travel characteristics are determined; Based on the trip characteristics and the trip database, determine whether the image set to be checked contains any anomalies: When the set of images to be verified contains an anomaly, the set of images to be verified will be marked as incorrect; When there are no anomalies in the image set to be verified, the first land cover type in the area to be verified is determined according to the land cover database; Based on each of the photographs in the image set to be verified, determine the type of each second land feature within the photograph; Based on the first land cover type in the area to be verified and the second land cover type in each of the photos in the image set to be verified, the qualified image set to be verified is determined.

2. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 1, characterized in that, The step of determining the travel features based on the set of images to be verified includes: Based on the image set to be verified, determine the first coordinate information and the absolute timestamp of each photo in the image set to be verified. The path order is determined based on the first coordinate information and the trip database; Based on the path order and each absolute timestamp, determine the time interval between two adjacent absolute timestamps in each absolute timestamp; Determine the time proportion based on each of the aforementioned time intervals; Based on the path order and each of the first coordinate information, determine the interval distance between two adjacent first coordinate information in each of the first coordinate information; Determine the distance ratio based on each of the aforementioned intervals; Determine whether the time ratio and the distance ratio are consistent; When the time ratio and the distance ratio are inconsistent, the image set to be checked will be labeled as incorrect. When the time ratio and the distance ratio are consistent, each of the interval time periods is used as the travel feature.

3. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 2, characterized in that, The step of determining whether the image set to be checked has any anomalies based on the trip features and the trip database includes: Based on the first coordinate information, determine the region information of the image set to be verified; Based on the regional information and the first coordinate information, determine the traffic attributes of each travel feature; Based on the aforementioned travel database and regional information, several movement characteristics are obtained; Determine whether the traffic attributes and movement speed of the travel characteristics simultaneously meet one of the movement characteristics; When both the traffic attribute and movement speed of the travel characteristics meet any one of the movement characteristics, the following step is executed: If there are no anomalies in the image set to be verified, the step of determining the first land cover type in the area to be verified based on the land cover database is executed: When neither the traffic attribute nor the speed of movement of the trip features matches any of the movement features, the step of labeling the image set to be checked as incorrect is executed.

4. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 2, characterized in that, The step of labeling the image set to be verified as incorrect when the time ratio and the distance ratio are inconsistent includes: When the time ratio and the distance ratio are inconsistent, determine the ratio result; Based on the similarity of the ratio results, several ratio groups are determined; Determine whether the number of each ratio group is greater than a first preset number; When the number of each ratio group is greater than the first preset number, the step of labeling the image set to be checked as incorrect is executed. When the number of each ratio group is less than or equal to the first preset number, several groups are determined according to each ratio group and each time interval. Each of the aforementioned groups is respectively identified as the travel feature.

5. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 4, characterized in that, The step of determining whether the image set to be checked has any anomalies based on the trip features and the trip database includes: Based on the aforementioned travel database, several movement features are obtained; Determine whether each of the aforementioned travel features contains a matching movement feature; When all the aforementioned travel features match the specified movement features, the following step is executed: if the image set to be verified does not contain any anomalies, the step of determining the first land cover type in the area to be verified based on the land cover database is performed: If any of the described travel features does not match the described movement feature, the step of labeling the image set to be checked as incorrect is executed.

6. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in any one of claims 2-5, characterized in that, The step of determining the type of each second land feature within each photograph in the image set to be verified includes: Based on the set of images to be verified, determine the pixel blocks in each of the photos in the set of images to be verified; Obtain confirmation information, and determine the type of each second land feature in the photo based on the confirmation information and each pixel block.

7. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 6, characterized in that, The step of determining a qualified image set to be verified based on each first land feature type in the area to be verified and each second land feature type in each photograph of the image set to be verified includes: Based on each first land feature type in the area to be verified and each second land feature type in each photo in the image set to be verified, determine whether the number of similarities between each second land feature type and each first land feature type is greater than a second preset number; When the number of similarities between each of the second land cover types and each of the first land cover types is greater than or equal to the second preset number, the step of determining the qualified set of images to be checked is executed; When the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number, the image set to be verified is marked as pending verification.

8. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 7, characterized in that, The step of marking the image set to be verified as pending verification when the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number includes: When the number of similarities between each of the second land cover types and each of the first land cover types is less than the second preset number, the photo is marked as abnormal; Determine whether the ratio of the photos marked as abnormal to the sum of the photos in the image set to be checked is greater than a preset ratio; When the ratio of the photos marked as abnormal to the sum of the photos in the image set to be verified is greater than or equal to the preset ratio, the step of marking the image set to be verified as pending verification is executed. When the ratio of the photos marked as abnormal to the sum of the photos in the image set to be verified is less than the preset ratio, the step of determining the qualified image set to be verified is performed.

9. The intelligent verification method for land use change evidence photos based on multi-source information fusion as described in claim 7, characterized in that, The step of determining the path order based on each of the first coordinate information and the trip database includes: Based on the land category database, determine the coordinate region within the area to be verified; Determine whether each of the first coordinate information is located within the coordinate region; When all the first coordinate information is located within the coordinate area, the path order is determined based on each of the first coordinate information and the trip database; When any of the first coordinate information is outside the coordinate region, the step of labeling the image set to be checked as incorrect is executed.

10. A smart verification system for land use change evidence photos based on multi-source information fusion, characterized in that: The method for intelligent verification of land change evidence photos based on multi-source information fusion as described in any one of claims 1-9 includes an execution module, a judgment module, and a calculation module; The execution module is used to acquire the image set to be verified, the travel database, and the land category database respectively; and to determine the travel characteristics based on the image set to be verified and the travel database. The judgment module is used to: determine whether the image set to be checked is abnormal based on the trip features and the trip database; when the image set to be checked is abnormal, the image set to be checked is marked as incorrect; When there are no anomalies in the image set to be verified, the first land cover type in the area to be verified is determined according to the land cover database; The calculation module is used to determine the type of each second land feature in each of the photos in the image set to be verified; Based on the first land cover type in the area to be verified and the second land cover type in each of the photos in the image set to be verified, the qualified image set to be verified is determined.

Images