An artificial intelligence-assisted system for creating and analyzing 3D maps

Abstract

The present invention is a system that converts aerial photographs into 3D maps using advanced image processing techniques and artificial intelligence algorithms, characterized in that it comprises the process steps of obtaining high-resolution images using unmanned aerial vehicles or airplanes with a high-resolution camera (100), using image processing techniques to reduce noise through filtering means, stabilize contrast and brightness, and correct faulty pixels (200), improving image quality with artificial intelligence techniques (300), and creating 3D models by combining images taken with photogrammetry software (400).

Description

[0001] AN ARTIFICIAL INTELLIGENCE-ASSISTED SYSTEM FOR CREATING AND ANALYZING 3D MAPS

[0002] Technical Field:

[0003] The present invention is a system that converts aerial photographs into 3D maps using advanced image processing techniques and artificial intelligence algorithms.

[0004] State of the Art:

[0005] Traditional 3D mapping and modeling methods are often costly, time-consuming, and highly user-dependent. Existing methods, used especially for mapping large areas or complex structures, face problems such as limited resolution and accuracy. These challenges make it difficult to collect and process critical data in areas such as urban planning, agricultural analysis, and natural disaster management.

[0006] In the present art, 3D mapping and modeling is usually performed by technologies such as photogrammetry, LIDAR (Light Detection and Ranging), and conventional satellite imaging. Photogrammetry extracts the 3D structure of objects by analyzing a series of photographs, while LIDAR technology collects distance and elevation information using laser signals. Satellite imaging is used for mapping large areas.

[0007] Although these methods are capable of large-scale data collection and analysis, they have complex processes requiring manual intervention and high costs. For example, in photogrammetric analysis, processing photographs and creating 3D models is a timeconsuming process and requires expertise. Although LIDAR offers high precision, equipment costs are high and data processing requires a lot of computing power.

[0008] The patent application no. TR201802566 describes a 3D map. This invention is related to a 3D map that can give a sense of realism and allow the observation of geographical effects during the study of historical events, snapshots or a specific scenario. The aim of the invention is to develop a 3D map that enables 3D map users to understand how historical events develop with geographical effects, and a projection that shows related animations and graphics on this map. The invention also aims to develop a management panel for content management specific to the surface on which animation and graphic content is displayed.

[0009] The application described above is not about 3D map modeling, but aims to understand nature and human movements, to examine the change of the map in the course of history, and to make inferences about how the earth is shaped according to geographical conditions.

[0010] The patent application no. TR201821313 describes the production of three-dimensional maps for indoor spaces with the use of geolocation data. The invention relates to a three-dimensional mapping method that is used for three-dimensional mapping of many indoor spaces such as shopping malls, business centers, museums, hospitals, universities, parking lots, and even mines, which provides compatibility and seamless integration with all spatial technologies, and which introduces this with the use of geographical location data such as latitude, longitude, and elevation.

[0011] The above-mentioned invention relates to 3D map modeling. Said application relates solely to obtaining geolocation data for enclosed spaces using latitude, longitude, and elevation theories of the relevant landscapes. However, photography techniques are not used in the application, and it is clear that a realistic 3D model cannot be created with said invention.

[0012] Definition of the Invention:

[0013] The present invention has a wide range of technical interest and uses, and offers innovative solutions in various sectors. In particular, it can be used in all areas that require highly accurate 3D mapping and modeling. The technical relevance of the invention is closely linked to advanced technologies such as artificial intelligence, image processing, photogrammetry, and data analytics.

[0014] The main technical problem that led to the emergence of this project is the need to make highly accurate and detailed 3D mapping processes more accessible, fast, and automated. Furthermore, the integration of image processing techniques and artificial intelligence algorithms aims to improve data accuracy, minimize errors caused by manual data processing processes, and speed up the processing of large data sets. Since the invention aims to provide innovative solutions and increase operational efficiency in a wide range of technical fields, especially by making complex geographical data understandable, it can be used in fields such as urban planning and infrastructure projects, agriculture and forest management, natural disaster management, architecture and construction sector, geology and mining, defense and security, environmental protection and natural disaster management, archaeology, etc.

[0015] Another aim of the invention is to make this process more accessible and cost-effective by eliminating the challenges posed by traditional methods. This system with an automated structure minimizes human intervention, improves data accuracy, and enables fast processing of large data sets.

[0016] The invention improves data accuracy by using high-resolution photos and advanced algorithms. Thanks to the optimization of images with artificial intelligence, it enables the creation of more precise and detailed 3D models.

[0017] The invention allows for similarly accurate results at lower costs, increasing usability in projects of different scales. It also reduces costs with a structure based on the processing of commonly used aerial photographs.

[0018] Description of the Drawings:

[0019] The invention will be described with reference to the accompanying figures, so that the features of the invention will be more clearly understood and appreciated, however it is not intended to limit the invention to these particular embodiments. On the contrary, it is intended all alternatives, modifications, and equivalences that may be included in the field of the invention as defined by the accompanying claims are within the scope. It should be understood that the details shown are for the sole purpose of illustrating preferred embodiments of the present invention and are intended to provide the most useful and easily understandable description of both the embodiment of the methods and the rules and conceptual features of the invention. In the drawings;

[0020] Fig. 1 is a schematic diagram showing the operation of the invention. The figures which will help understand this invention are numbered as indicated in the accompanying drawing and are given below with their names.

[0021] Description of the References:

[0022] 100. Obtaining images

[0023] 200. Using image processing techniques

[0024] 300. Improving image quality

[0025] 400. Creating 3D model

[0026] Detailed Description of the Invention:

[0027] The methods related to the process of the invention are respectively; obtaining images (100), using image processing techniques (200), improving image quality (300), creating 3D model (400).

[0028] For the 3D model, data entry is required first. For the initial data input, drones or airplanes will be used to acquire high-resolution photographs (100) from large areas. For this, high-resolution cameras are used. The location information and elevation information of the obtained photographs will also be obtained via GPS. Once obtained, the images are then processed using image processing techniques (200) to reduce noise and stabilize contrast and brightness. In the process, missing data is completed and faulty pixels are corrected. Low pass filters, bilateral filters, etc. can be used for noise reduction. Furthermore, deep learning models and HDR techniques can be used to reduce noise and optimize contrast and brightness. After image processing techniques, methods such as edge detection, object recognition, deep learning model, etc. are used to improve the quality of the images (300), making them most suitable for 3D modeling. Human intervention in this process is minimal. Finally, aerial photographs taken from various angles are combined using photogrammetry software and a 3D model is created (400).

Claims

CLAIMS1 . An artificial intelligence-assisted system for creating and analyzing 3D maps, characterized in that it comprises the process steps of: - obtaining high-resolution images using unmanned aerial vehicles or airplanes with a high-resolution camera (100),- using image processing techniques to reduce noise through filtering means, stabilize contrast and brightness, and correct faulty pixels (200),- improving image quality with artificial intelligence techniques (300), - creating 3D models by combining images taken with photogrammetry software (400).

Images