DUDMap: 3D RGB-D mapping for dense, unstructured, and dynamic environment

2021 ◽  
Vol 18 (3) ◽  
pp. 172988142110161
Author(s):  
Özgür Hastürk ◽  
Aydan M Erkmen
Keyword(s):  
Real Time ◽  
Dynamic Environment ◽  
Semantic Segmentation ◽  
Small Region ◽  
Dynamic Environments ◽  
Reconstruction Error ◽  
Localization And Mapping ◽  
Segmentation Methods ◽  
High Dynamic ◽  
Better Than

Simultaneous localization and mapping (SLAM) problem has been extensively studied by researchers in the field of robotics, however, conventional approaches in mapping assume a static environment. The static assumption is valid only in a small region, and it limits the application of visual SLAM in dynamic environments. The recently proposed state-of-the-art SLAM solutions for dynamic environments use different semantic segmentation methods such as mask R-CNN and SegNet; however, these frameworks are based on a sparse mapping framework (ORBSLAM). In addition, segmentation process increases the computational power, which makes these SLAM algorithms unsuitable for real-time mapping. Therefore, there is no effective dense RGB-D SLAM method for real-world unstructured and dynamic environments. In this study, we propose a novel real-time dense SLAM method for dynamic environments, where 3D reconstruction error is manipulated for identification of static and dynamic classes having generalized Gaussian distribution. Our proposed approach requires neither explicit object tracking nor object classifier, which makes it robust to any type of moving object and suitable for real-time mapping. Our method eliminates the repeated views and uses consistent data that enhance the performance of volumetric fusion. For completeness, we compare our proposed method using different types of high dynamic dataset, which are publicly available, to demonstrate the versatility and robustness of our approach. Experiments show that its tracking performance is better than other dense and dynamic SLAM approaches.

scholarly journals SLAM in Dynamic Environments: A Deep Learning Approach for Moving Object Tracking Using ML-RANSAC Algorithm

Sensors ◽  
2019 ◽  
Vol 19 (17) ◽  
pp. 3699 ◽  
Author(s):  
Masoud S. Bahraini ◽  
Ahmad B. Rad ◽  
Mohammad Bozorg
Keyword(s):  
Target Tracking ◽  
Real Time ◽  
Dynamic Environment ◽  
Dynamic Environments ◽  
Depth Information ◽  
Localization And Mapping ◽  
Vision Sensors ◽  
Multi Target Tracking ◽  
Stationary Objects ◽  
Feature Based

The important problem of Simultaneous Localization and Mapping (SLAM) in dynamic environments is less studied than the counterpart problem in static settings. In this paper, we present a solution for the feature-based SLAM problem in dynamic environments. We propose an algorithm that integrates SLAM with multi-target tracking (SLAMMTT) using a robust feature-tracking algorithm for dynamic environments. A novel implementation of RANdomSAmple Consensus (RANSAC) method referred to as multilevel-RANSAC (ML-RANSAC) within the Extended Kalman Filter (EKF) framework is applied for multi-target tracking (MTT). We also apply machine learning to detect features from the input data and to distinguish moving from stationary objects. The data stream from LIDAR and vision sensors are fused in real-time to detect objects and depth information. A practical experiment is designed to verify the performance of the algorithm in a dynamic environment. The unique feature of this algorithm is its ability to maintain tracking of features even when the observations are intermittent whereby many reported algorithms fail in such situations. Experimental validation indicates that the algorithm is able to perform consistent estimates in a fast and robust manner suggesting its feasibility for real-time applications.

scholarly journals A Real-Time Structure of Attitude Algorithm for High Dynamic Bodies

2017 ◽  
Vol 2017 ◽  
pp. 1-8
Author(s):  
Xingcheng Li ◽  
Shuangbiao Zhang
Keyword(s):  
Real Time ◽  
Time Structure ◽  
Flow Diagram ◽  
The Real ◽  
Time Problem ◽  
Conventional Structure ◽  
High Dynamic ◽  
Two Stages ◽  
Better Than ◽  
Attitude Matrix

To solve the real-time problem of attitude algorithm for high dynamic bodies, a real-time structure of attitude algorithm is developed by analyzing the conventional structure that has two stages, and a flow diagram of a real-time structure for a Matlab program is provided in detail. During the update of the attitude matrix, the real-time structure saves every element of attitude matrix in minor loop in real time and updates the next attitude matrix based on the previous matrix every subsample time. Thus, the real-time structure avoids lowering updating frequency, though the multisubsample algorithms are used. Simulation and analysis show that the real-time structure of attitude algorithm is better than the conventional structure due to short update time of attitude matrix and small drifting error, and it is more appropriate for high dynamic bodies.

Multi-Robot SLAM in Dynamic Environments with Parallel Maps

2021 ◽  
pp. 2150011
Author(s):  
Sajad Badalkhani ◽  
Ramazan Havangi ◽  
Mohsen Farshad
Keyword(s):  
Large Scale ◽  
Dynamic Environment ◽  
Dynamic Environments ◽  
Extensive Literature ◽  
Real World Data ◽  
Data Set ◽  
Cooperative Approach ◽  
Localization And Mapping ◽  
Multi Robot

There is an extensive literature regarding multi-robot simultaneous localization and mapping (MRSLAM). In most part of the research, the environment is assumed to be static, while the dynamic parts of the environment degrade the estimation quality of SLAM algorithms and lead to inherently fragile systems. To enhance the performance and robustness of the SLAM in dynamic environments (SLAMIDE), a novel cooperative approach named parallel-map (p-map) SLAM is introduced in this paper. The objective of the proposed method is to deal with the dynamics of the environment, by detecting dynamic parts and preventing the inclusion of them in SLAM estimations. In this approach, each robot builds a limited map in its own vicinity, while the global map is built through a hybrid centralized MRSLAM. The restricted size of the local maps, bounds computational complexity and resources needed to handle a large scale dynamic environment. Using a probabilistic index, the proposed method differentiates between stationary and moving landmarks, based on their relative positions with other parts of the environment. Stationary landmarks are then used to refine a consistent map. The proposed method is evaluated with different levels of dynamism and for each level, the performance is measured in terms of accuracy, robustness, and hardware resources needed to be implemented. The method is also evaluated with a publicly available real-world data-set. Experimental validation along with simulations indicate that the proposed method is able to perform consistent SLAM in a dynamic environment, suggesting its feasibility for MRSLAM applications.

scholarly journals Real-Time Semantic Segmentation with Dual Encoder and Self-Attention Mechanism for Autonomous Driving

Sensors ◽  
2021 ◽  
Vol 21 (23) ◽  
pp. 8072
Author(s):  
Yu-Bang Chang ◽  
Chieh Tsai ◽  
Chang-Hong Lin ◽  
Poki Chen
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Network Architecture ◽  
Semantic Segmentation ◽  
Autonomous Driving ◽  
Attention Mechanism ◽  
Trade Off ◽  
Segmentation Methods ◽  
General Semantic ◽  
Deep Learning Model

As the techniques of autonomous driving become increasingly valued and universal, real-time semantic segmentation has become very popular and challenging in the field of deep learning and computer vision in recent years. However, in order to apply the deep learning model to edge devices accompanying sensors on vehicles, we need to design a structure that has the best trade-off between accuracy and inference time. In previous works, several methods sacrificed accuracy to obtain a faster inference time, while others aimed to find the best accuracy under the condition of real time. Nevertheless, the accuracies of previous real-time semantic segmentation methods still have a large gap compared to general semantic segmentation methods. As a result, we propose a network architecture based on a dual encoder and a self-attention mechanism. Compared with preceding works, we achieved a 78.6% mIoU with a speed of 39.4 FPS with a 1024 × 2048 resolution on a Cityscapes test submission.

scholarly journals Skin lesion segmentation method for dermoscopic images with convolutional neural networks and semantic segmentation

2021 ◽  
Vol 45 (1) ◽  
pp. 122-129
Author(s):  
Dang N.H. Thanh ◽  
Nguyen Hoang Hai ◽  
Le Minh Hieu ◽  
Prayag Tiwari ◽  
V.B. Surya Prasath
Keyword(s):  
Skin Cancer ◽  
Skin Lesion ◽  
Semantic Segmentation ◽  
Segmentation Method ◽  
Lesion Segmentation ◽  
Computing Systems ◽  
Segmentation Methods ◽  
Dermoscopic Image ◽  
Better Than

Melanoma skin cancer is one of the most dangerous forms of skin cancer because it grows fast and causes most of the skin cancer deaths. Hence, early detection is a very important task to treat melanoma. In this article, we propose a skin lesion segmentation method for dermoscopic images based on the U-Net architecture with VGG-16 encoder and the semantic segmentation. Base on the segmented skin lesion, diagnostic imaging systems can evaluate skin lesion features to classify them. The proposed method requires fewer resources for training, and it is suitable for computing systems without powerful GPUs, but the training accuracy is still high enough (above 95 %). In the experiments, we train the model on the ISIC dataset – a common dermoscopic image dataset. To assess the performance of the proposed skin lesion segmentation method, we evaluate the Sorensen-Dice and the Jaccard scores and compare to other deep learning-based skin lesion segmentation methods. Experimental results showed that skin lesion segmentation quality of the proposed method are better than ones of the compared methods.

scholarly journals A Semantic SLAM System for Catadioptric Panoramic Cameras in Dynamic Environments

Sensors ◽  
2021 ◽  
Vol 21 (17) ◽  
pp. 5889
Author(s):  
Yu Zhang ◽  
Xiping Xu ◽  
Ning Zhang ◽  
Yaowen Lv
Keyword(s):  
Dynamic Environment ◽  
Dynamic Environments ◽  
Data Sets ◽  
Panoramic Image ◽  
Feature Points ◽  
Moving Targets ◽  
Panoramic Cameras ◽  
Public Data ◽  
Dynamic Objects ◽  
Better Than

When a traditional visual SLAM system works in a dynamic environment, it will be disturbed by dynamic objects and perform poorly. In order to overcome the interference of dynamic objects, we propose a semantic SLAM system for catadioptric panoramic cameras in dynamic environments. A real-time instance segmentation network is used to detect potential moving targets in the panoramic image. In order to find the real dynamic targets, potential moving targets are verified according to the sphere’s epipolar constraints. Then, when extracting feature points, the dynamic objects in the panoramic image are masked. Only static feature points are used to estimate the pose of the panoramic camera, so as to improve the accuracy of pose estimation. In order to verify the performance of our system, experiments were conducted on public data sets. The experiments showed that in a highly dynamic environment, the accuracy of our system is significantly better than traditional algorithms. By calculating the RMSE of the absolute trajectory error, we found that our system performed up to 96.3% better than traditional SLAM. Our catadioptric panoramic camera semantic SLAM system has higher accuracy and robustness in complex dynamic environments.

scholarly journals Circle-U-Net: An Efficient Architecture for Semantic Segmentation

Algorithms ◽  
2021 ◽  
Vol 14 (6) ◽  
pp. 159
Author(s):  
Feng Sun ◽  
Ajith Kumar V ◽  
Guanci Yang ◽  
Ansi Zhang ◽  
Yiyun Zhang
Keyword(s):  
State Of The Art ◽  
Semantic Segmentation ◽  
Proposed Model ◽  
Segmentation Methods ◽  
Deep Networks ◽  
Improved Accuracy ◽  
Better Than

State-of-the-art semantic segmentation methods rely too much on complicated deep networks and thus cannot train efficiently. This paper introduces a novel Circle-U-Net architecture that exceeds the original U-Net on several standards. The proposed model includes circle connect layers, which is the backbone of ResUNet-a architecture. The model possesses a contracting part with residual bottleneck and circle connect layers that capture context and expanding paths, with sampling layers and merging layers for a pixel-wise localization. The results of the experiment show that the proposed Circle-U-Net achieves an improved accuracy of 5.6676%, 2.1587% IoU (Intersection of union, IoU) and can detect 67% classes greater than U-Net, which is better than current results.

scholarly journals MBFFNet: Multi-Branch Feature Fusion Network for Colonoscopy

2021 ◽  
Vol 9 ◽  
Author(s):  
Houcheng Su ◽  
Bin Lin ◽  
Xiaoshuang Huang ◽  
Jiao Li ◽  
Kailin Jiang ◽  
...  
Keyword(s):  
Real Time ◽  
Feature Fusion ◽  
Rapid Development ◽  
Semantic Segmentation ◽  
Medical Image Segmentation ◽  
Feature Maps ◽  
Improve Model ◽  
Segmentation Methods ◽  
Rectal Polyps ◽  
Good Potential

Colonoscopy is currently one of the main methods for the detection of rectal polyps, rectal cancer, and other diseases. With the rapid development of computer vision, deep learning–based semantic segmentation methods can be applied to the detection of medical lesions. However, it is challenging for current methods to detect polyps with high accuracy and real-time performance. To solve this problem, we propose a multi-branch feature fusion network (MBFFNet), which is an accurate real-time segmentation method for detecting colonoscopy. First, we use UNet as the basis of our model architecture and adopt stepwise sampling with channel multiplication to integrate features, which decreases the number of flops caused by stacking channels in UNet. Second, to improve model accuracy, we extract features from multiple layers and resize feature maps to the same size in different ways, such as up-sampling and pooling, to supplement information lost in multiplication-based up-sampling. Based on mIOU and Dice loss with cross entropy (CE), we conduct experiments in both CPU and GPU environments to verify the effectiveness of our model. The experimental results show that our proposed MBFFNet is superior to the selected baselines in terms of accuracy, model size, and flops. mIOU, F score, and Dice loss with CE reached 0.8952, 0.9450, and 0.1602, respectively, which were better than those of UNet, UNet++, and other networks. Compared with UNet, the flop count decreased by 73.2%, and the number of participants also decreased. The actual segmentation effect of MBFFNet is only lower than that of PraNet, the number of parameters is 78.27% of that of PraNet, and the flop count is 0.23% that of PraNet. In addition, experiments on other types of medical tasks show that MBFFNet has good potential for general application in medical image segmentation.

Sign in / Sign up

Export Citation Format

Share Document