A Deep Learning-Based Semantic Filter for RANSAC-Based Fundamental Matrix Calculation and the ORB-SLAM System

Shao, Chunyan; Zhang, Chi; Fang, Zaojun; Yang, Guilin

doi:10.1109/access.2019.2962268

Cited by 16 publications

(14 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A vision-based deep learning network can be used to detect the environment for conducting motion detection, walkable area detection, and motion planning [19]. Obstacle detection can also performed by object detection, and the technique algorithm will be used to mark walkable areas and relative coordinates [20][21].…”

Section: Literature Review and Methodologymentioning

confidence: 99%

Vehicle Path Planning with Multicloud Computation Services

Wang

Lin

Sheu

2021

Adv. technol. innov.

View full text Add to dashboard Cite

With the development of artificial intelligence, public cloud service platforms have begun to provide common pretrained object recognition models for public use. In this study, a dynamic vehicle path-planning system is developed, which uses several general pretrained cloud models to detect obstacles and calculate the navigation area. The Euclidean distance and the inequality based on the detected marker box data are used for vehicle path planning. Experimental results show that the proposed method can effectively identify the driving area and plan a safe route. The proposed method integrates the bounding box information provided by multiple cloud object detection services to detect navigable areas and plan routes. The time required for cloud-based obstacle identification is 2 s per frame, and the time required for feasible area detection and action planning is 0.001 s per frame. In the experiments, the robot that uses the proposed navigation method can plan routes successfully.

show abstract

Section: Literature Review and Methodologymentioning

confidence: 99%

Vehicle Path Planning with Multicloud Computation Services

Wang

Lin

Sheu

2021

Adv. technol. innov.

View full text Add to dashboard Cite

show abstract

“…Zheng [30] proposes an image registration method by using structured topological constraints, which establishes the correspondences of local points, triangular edges and triangular surfaces to dynamically erase mismatches. Recently, a semantic filter based on faster R-CNN is proposed to integrate with RANSAC [31], and three bad labels denoting low-quality feature correspondences are filtered out during the pre-processing phase. As shown in Figure .2, the wrong matches want to be removed, while the correct matches want to be retained after the prefiltering.…”

Section: Related Workmentioning

confidence: 99%

Accelerated RANSAC for Accurate Image Registration in Aerial Video Surveillance

et al. 2021

View full text Add to dashboard Cite

Compared with ground views and direct overhead views (for orbital satellites), aerial robotics allow for capturing videos from diverse viewpoints and scenes, thus, the content of aerial image is complex and changeable, and aerial video has complex inter-frame transforms stemming from the blend of camera motion, platform motion and jitter. In addition, poor quality and similar texture are common in longdistance and large-scale aerial video surveillance. All of these interferences make image registration of aerial video difficult. This paper puts forward one image registration method suited to aerial video on the basis of the hypothesize-and-verify of RANSAC. The proposed accelerated RANSAC, named PSSC-RANSAC (Prior Sampling & Sample Check RANSAC), incorporates prior sampling, which comes from three levels of sample evaluation, including texture magnitude, spatial consistency and feature similarity, to generate more possibly correct samples in priority. Furthermore, prior information of sample, quality of sample subset and subset invariability are together used to check the sample subsets, and the incompatible arrangements of subsets are immediately ruled out in sample check stage, which speeds up the iteration further. Results of the experiment have proved the good performance of the presented PSSC-RANSAC at 90% contamination level. For typical image pairs, the number of iterations is reduced by at least 16.67% and evaluation computation is reduced by at least 11.01% compared with SVH-RANSAC, and the reprojection error is decreased by at least 4.44% and 6.31% compared with RANSAC and SVH-RANSAC, respectively. It can overcome the interferences, and is very suitable for image registration of aerial images.

show abstract

“…Due to the good performance of the ORB characteristics [10][11][12][13][14][15], many scholars at home and abroad have made different improvements to the ORB algorithm [16][17][18][19]. Hong et al [20] matched the ORB feature point matching algorithm and eight parameters and combined the rotation model, improving the detection speed of the feature point; and Bing et al [21] improved the rotation of the ORB algorithm in the feature point matching algorithm of the ORB, which enhanced the matching accuracy; but for special scenes, the study of the ORB algorithm is rare under the conditions of poor illumination conditions.…”

Section: Introductionmentioning

confidence: 99%

Fast Target Recognition Based on Improved ORB Feature

et al. 2022

View full text Add to dashboard Cite

A novel fast target recognition algorithm is proposed under the dynamic scene moving target recognition. Aiming at the poor matching effect of the traditional Oriented Fast and Rotated Brief (ORB) algorithm on underexposed or overexposed images caused by illumination, the idea of combining adaptive histogram equalization with the ORB algorithm is proposed to get better feature point quality and matching efficiency. First, the template image and each frame of the video stream are processed by grayscale. Second, the template image and the image to be input in the video stream are processed by adaptive histogram equalization. Third, the feature point descriptors of the ORB feature are quantized by the Hamming distance. Finally, the K-nearest-neighbor (KNN) matching algorithm is used to match and screen feature points. According to the matching good feature point logarithm, a reasonable threshold is established and the target is classified. The comparison and verification are carried out by experiments. Experimental results show that the algorithm not only maintains the superiority of ORB itself but also significantly improves the performance of ORB under the conditions of underexposure or overexposure. The matching effect of the image is robust to illumination, and the target to be detected can be accurately identified in real time. The target can be accurately classified in the small sample scene, which can meet the actual production requirements.

show abstract

A Deep Learning-Based Semantic Filter for RANSAC-Based Fundamental Matrix Calculation and the ORB-SLAM System

Cited by 16 publications

References 39 publications

Vehicle Path Planning with Multicloud Computation Services

Vehicle Path Planning with Multicloud Computation Services

Accelerated RANSAC for Accurate Image Registration in Aerial Video Surveillance

Fast Target Recognition Based on Improved ORB Feature

Contact Info

Product

Resources

About