Multi-Modal Scene Matching Location Algorithm Based on M2Det
Source abstract
Abstract
In recent years, many visual positioning algorithms have been proposed based on computer vision and they have achieved good results. However, these algorithms have a single function, cannot perceive the environment, and have poor versatility, and there is a certain mismatch phenomenon, which affects the positioning accuracy. Therefore, this paper proposes a location algorithm that combines a target recognition algorithm with a depth feature matching algorithm to solve the problem of unmanned aerial vehicle (UAV) environment perception and multi-modal image-matching fusion location. This algorithm was based on the single-shot object detector based on multi-level feature pyramid network (M2Det) algorithm and replaced the original visual geometry group (VGG) feature extraction network with the ResNet-101 network to improve the feature extraction capability of the network model. By introducing a depth feature matching algorithm, the algorithm shares neural network weights and realizes the design of UAV target recognition and a multi-modal image-matching fusion positioning algorithm. When the reference image and the real-time image were mismatched, the dynamic adaptive proportional constraint and the random sample consensus consistency algorithm (DAPC-RANSAC) were used to optimize the matching results to improve the correct matching efficiency of the target. Using the multi-modal registration data set, the proposed algorithm was compared and analyzed to verify its superiority and feasibility. The results show that the algorithm proposed in this paper can effectively deal with the matching between multi-modal images (visible image–infrared image, infrared image–satellite image, visible image–satellite image), and the contrast, scale, brightness, ambiguity deformation, and other changes had good stability and robustness. Finally, the effectiveness and practicability of the algorithm proposed in this paper were verified in an aerial test scene of an S1000 six-rotor UAV.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Robotics and Sensor-Based Localization · Advanced Image and Video Retrieval Techniques · Infrared Target Detection Methodologies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (65)
Show source-provided references
1. Research on the applications of Russian precision strike weapons in the Russia-Ukraine conflict
2. UAV in the advent of the twenties: Where we stand and what is next10.1016/j.isprsjprs.2021.12.006 ↗
3. Covert spoofing algorithm of UAV based on GPS/INS integrated navigation10.1109/TVT.2019.2914477 ↗
4. Collision-free autonomous navigation of a small UAV using low-cost sensors in GPS-denied environments10.1007/s12555-019-0797-7 ↗
5. Memory-enhanced cognitive radar for autonomous navigation10.1049/iet-rsn.2019.0409 ↗
6. Enabling robust and accurate navigation for UAVs using real-time GNSS precise point positioning and IMU integration10.1017/aer.2020.80 ↗
7. Robust GNSS-denied localization for UAV using particle filter and visual odometry10.1007/s00138-019-01046-4 ↗
8. End-to-end UAV simulation for visual SLAM and navigation
9. A simple and robust feature point matching algorithm based on restricted spatial order constraints for aerial image registration10.1109/TGRS.2011.2160645 ↗
10. Image-based Monte-Carlo localization with information allocation logic to mitigate shadow effect10.1109/ACCESS.2020.3039413 ↗
11. Assisting UAV localization via deep contextual image-matching10.1109/JSTARS.2021.3054832 ↗
12. A practical cross-view image-matching method between UAV and satellite for UAV-based geo-localization10.3390/rs13010047 ↗
13. A review of multimodal image-matching: Methods and applications10.1016/j.inffus.2021.02.012 ↗
14. Registration of UAV images using improved structural shape similarity based on mathematical morphology and phase congruency10.1109/JSTARS.2020.2982929 ↗
15. Fast unmanned aerial vehicle image-matching combining geometric information and feature similarity
16. A new productive framework for point-based matching of airplane oblique and UAV based images10.1111/phor.12374 ↗
17. Automatic matching of multimodal remote sensing images via learned unstructured road feature
18. Distinctive image features from scale-invariant keypoints10.1023/B:VISI.0000029664.99615.94 ↗
19. ORB: An efficient alternative to SIFT or SURF
20. Fast scene matching method based on scale invariant feature transform
21. Fingerprint liveness detection using histogram of oriented gradient based texture feature
22. Artwork painting identification method for panorama based on adaptive rectilinear projection and optimized ASIFT10.1007/s11042-019-07985-4 ↗
23. FSD-BRIEF: A distorted BRIEF descriptor for fisheye image based on spherical perspective model
24. RIFT: Multi-modal image-matching based on radiation-variation insensitive feature transform10.1109/TIP.2019.2959244 ↗
25. LNIFT: Locally normalized image for rotation invariant multimodal feature matching
26. Selective transfer cycle GAN for unsupervised person re-identification10.1007/s11042-019-08604-y ↗
27. ClawGAN: Claw connection-based generative adversarial networks for facial image translation in thermal to RGB visible light10.1016/j.eswa.2021.116269 ↗
28. Dual distribution matching GAN10.1016/j.neucom.2021.12.095 ↗
29. Match ACNN: A multi-granularity deep matching model
30. Automatic detection of maize tassels from UAV images by combining random forest classifier and VGG1610.3390/rs12183049 ↗
31. Transfer learning for image classification using VGG19: Caltech-101 image data set
32. Deep residual learning for image recognition
33. A novel ResNet101 model based on dense dilated convolution for image classification10.1007/s42452-021-04897-7 ↗
34. Efficient densely connected convolutional neural networks
35. Lightweight residual densely connected convolutional neural network10.1007/s11042-020-09223-8 ↗
36. Single image dehazing via dual-path recurrent network10.1109/TIP.2021.3078319 ↗
37. A dual-path fusion network for pan-sharpening
38. Improved performance of image semantic segmentation using NASNet
39. MatchNet: Unifying feature and metric learning for patch-based matching
40. PPFNet: Global context aware local features for robust 3D point matching
41. LoFTR: Detector-free local feature matching with transformers
42. Discriminative learning of deep convolutional feature point descriptors
43. Learning to compare image patches via convolutional neural networks
44. Visible and infrared image-matching method based on generative adversarial model
45. Registration of multi-modal remote sensing image based on deep fully convolutional neural network10.1109/JSTARS.2019.2916560 ↗
46. D2-Net: A trainable cnn for joint description and detection of local features
47. R2D2: Reliable and repeatable detectors and descriptors for joint sparse keypoint detection and local feature extraction
48. SuperGlue: Learning feature matching with graph neural networks
49. Key.Net: Keypoint detection by handcrafted and learned CNN filters
50. V. Balntas, E. Johns and L. Tang, “PN-Net: Conjoined triple deep network for learning local image descriptors,” arXiv preprint arXiv: 1601.05030, 2016.
51. Fast and robust matching for multi-modal remote sensing image registration10.1109/TGRS.2019.2924684 ↗
52. A rotation-invariant optical and SAR image registration algorithm based on deep and gaussian features
53. Repeatability is not enough: Learning affine regions via discriminability
54. You only look once: Unified, real-time object detection
55. MAF-YOLO: Multi-modal attention fusion based YOLO for pedestrian detection10.1016/j.infrared.2021.103906 ↗
56. A real-time detection algorithm for kiwifruit defects based on YOLOv5
57. Local keypoint-based faster R-CNN10.1007/s10489-020-01665-9 ↗
58. Protein crystal instance segmentation based on mask R-CNN10.3390/cryst11020157 ↗
59. Object detection based on RGC mask R‐CNN10.1049/iet-ipr.2019.0057 ↗
60. Detection of cervical cells based on improved SSD network10.1007/s11042-021-11015-7 ↗
61. Recognizing human behaviors from surveillance videos using the SSD algorithm10.1007/s11227-020-03578-3 ↗
62. A fast SSD model based on parameter reduction and dilated convolution10.1007/s11554-021-01108-9 ↗
63. Ship detection from X-band SAR images using M2Det deep learning model10.3390/app10217751 ↗
64. Multi-scale detection for X-ray prohibited items in complex background
65. M2Det: A single-shot object detector based on multi-level feature pyramid network