Recognition and Classification of 3D Objects of Different Details
The development of 3D scanning technologies has made it possible to obtain an increasing amount of data about the external world, which leads to the need for efficient methods of processing acquired data to recognize objects. Traditional approaches face accuracy, speed, and reliability problems due...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Qubahan
2024-06-01
|
| Series: | Qubahan Academic Journal |
| Subjects: | |
| Online Access: | https://journal.qubahan.com/index.php/qaj/article/view/557 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832544589798440960 |
|---|---|
| author | Islam A. Alexandrov Maxim S. Mikhailov Alexander N. Muranov Vladimir Zh. Kuklin |
| author_facet | Islam A. Alexandrov Maxim S. Mikhailov Alexander N. Muranov Vladimir Zh. Kuklin |
| author_sort | Islam A. Alexandrov |
| collection | DOAJ |
| description |
The development of 3D scanning technologies has made it possible to obtain an increasing amount of data about the external world, which leads to the need for efficient methods of processing acquired data to recognize objects. Traditional approaches face accuracy, speed, and reliability problems due to the complexity and diversity of object shapes, sizes, degrees of detail, and the presence of noise and artifacts in the data. Therefore, our goal was to improve the object recognition efficiency. It is necessary to determine the method of obtaining geometric and topological parameters. In the paper it is proposed to use the method of Laplace-Beltrami, which allows to calculate distances and angles between points within a given area. Next, it is necessary to determine which parameters will be used to analyze the obtained geometric data. We propose the use of three spectral descriptors – Heat Kernel Signature (HKS), Weave Kernel Signature (WKS) and the wavelet descriptor (SGWT). Then, we develop a high-accuracy recognition method based on spectral and topological invariants processed using a convolutional neural network. Subsequently, the parameters of the descriptors are calculated, and then they are calculated through the neural network, resulting in the classification of the object. In summary, the structure of the proposed method comprises the computation of the Laplace-Beltrami spectrum, the construction of spectral distribution maps, and the subsequent processing of this information using a neural network. After analyzing the results, we found that the proposed method has a recognition rate of 0.9 s and recognition accuracy of 97%. It was shown how much more effective the use of the three descriptors was compared to the use of each one individually. An example of object recognition using the proposed method was also given. The methodology outlined in this paper utilizes machine learning to achieve high levels of accuracy in the classification of different objects. Effectiveness of the proposed method in this study enhance existing recognition systems and open new opportunities for their application in various fields, including robotics, agriculture, and navigation. This research demonstrates considerable potential for further development and application in the field of agriculture, underscoring the continued necessity for research in this area
|
| format | Article |
| id | doaj-art-f9ba52443439435abf679db3dfe44c70 |
| institution | Kabale University |
| issn | 2709-8206 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | Qubahan |
| record_format | Article |
| series | Qubahan Academic Journal |
| spelling | doaj-art-f9ba52443439435abf679db3dfe44c702025-02-03T10:11:53ZengQubahanQubahan Academic Journal2709-82062024-06-014210.48161/qaj.v4n2a557557Recognition and Classification of 3D Objects of Different DetailsIslam A. Alexandrov0Maxim S. Mikhailov1Alexander N. Muranov2Vladimir Zh. Kuklin3Institute of Design-Technology Informatics, Russian Academy of Sciences, Vadkovsky Lane, 18., building 1A, Moscow, 127055, Russian FederationInstitute of Design-Technology Informatics, Russian Academy of Sciences, Vadkovsky Lane, 18., building 1A, Moscow, 127055, Russian FederationInstitute of Design-Technology Informatics, Russian Academy of Sciences, Vadkovsky Lane, 18., building 1A, Moscow, 127055, Russian FederationInstitute of Design-Technology Informatics, Russian Academy of Sciences, Vadkovsky Lane, 18., building 1A, Moscow, 127055, Russian Federation The development of 3D scanning technologies has made it possible to obtain an increasing amount of data about the external world, which leads to the need for efficient methods of processing acquired data to recognize objects. Traditional approaches face accuracy, speed, and reliability problems due to the complexity and diversity of object shapes, sizes, degrees of detail, and the presence of noise and artifacts in the data. Therefore, our goal was to improve the object recognition efficiency. It is necessary to determine the method of obtaining geometric and topological parameters. In the paper it is proposed to use the method of Laplace-Beltrami, which allows to calculate distances and angles between points within a given area. Next, it is necessary to determine which parameters will be used to analyze the obtained geometric data. We propose the use of three spectral descriptors – Heat Kernel Signature (HKS), Weave Kernel Signature (WKS) and the wavelet descriptor (SGWT). Then, we develop a high-accuracy recognition method based on spectral and topological invariants processed using a convolutional neural network. Subsequently, the parameters of the descriptors are calculated, and then they are calculated through the neural network, resulting in the classification of the object. In summary, the structure of the proposed method comprises the computation of the Laplace-Beltrami spectrum, the construction of spectral distribution maps, and the subsequent processing of this information using a neural network. After analyzing the results, we found that the proposed method has a recognition rate of 0.9 s and recognition accuracy of 97%. It was shown how much more effective the use of the three descriptors was compared to the use of each one individually. An example of object recognition using the proposed method was also given. The methodology outlined in this paper utilizes machine learning to achieve high levels of accuracy in the classification of different objects. Effectiveness of the proposed method in this study enhance existing recognition systems and open new opportunities for their application in various fields, including robotics, agriculture, and navigation. This research demonstrates considerable potential for further development and application in the field of agriculture, underscoring the continued necessity for research in this area https://journal.qubahan.com/index.php/qaj/article/view/5573D scanninginvariantsdescriptorsmachine learningconvolutional neural network |
| spellingShingle | Islam A. Alexandrov Maxim S. Mikhailov Alexander N. Muranov Vladimir Zh. Kuklin Recognition and Classification of 3D Objects of Different Details Qubahan Academic Journal 3D scanning invariants descriptors machine learning convolutional neural network |
| title | Recognition and Classification of 3D Objects of Different Details |
| title_full | Recognition and Classification of 3D Objects of Different Details |
| title_fullStr | Recognition and Classification of 3D Objects of Different Details |
| title_full_unstemmed | Recognition and Classification of 3D Objects of Different Details |
| title_short | Recognition and Classification of 3D Objects of Different Details |
| title_sort | recognition and classification of 3d objects of different details |
| topic | 3D scanning invariants descriptors machine learning convolutional neural network |
| url | https://journal.qubahan.com/index.php/qaj/article/view/557 |
| work_keys_str_mv | AT islamaalexandrov recognitionandclassificationof3dobjectsofdifferentdetails AT maximsmikhailov recognitionandclassificationof3dobjectsofdifferentdetails AT alexandernmuranov recognitionandclassificationof3dobjectsofdifferentdetails AT vladimirzhkuklin recognitionandclassificationof3dobjectsofdifferentdetails |