Efficient real-time loop closure detection using GMM and tree structure

LiDAR-based loop closure detection is a crucial part of realizing robust SLAM algorithms for intelligent vehicles with LiDAR sensors. Existing methods often reduce the keypoint dimension to encode the global descriptor, which sacrifices the freedom of loop detection and correction. Based on the 6-DOF rigid transformation property of spatial triangles, we propose an algorithm for extracting and describing 3D keypoints from high-resolution spinning LiDAR intensity images to encode triangle descriptors, termed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">intensity triangle descriptor (ITD)</i>. In comparison to the direct extraction of keypoints from the point cloud, the use of image-derived feature points provides additional photometric texture information and better handles uneven spatial density of the point cloud, which is advantageous in unstructured and geometrically degraded scenes. To enhance the stability of keypoints, the spatial positions of multi-frame image feature points are registered to a keyframe by an odometer for voxel downsampling and non-maximum suppression, with the objective of reducing unstable feature points. For high discrimination, the neighbor image patches of each vertex (keypoint) are aggregated to estimate a Gaussian mixture model (GMM) as the keypoint signature. An efficient two-stage loop closure detection method is then proposed for ITD, consisting of candidate retrieval based on triangle side lengths and vertex GMMs, followed by geometric verification of matched descriptor pairs. The effectiveness of the proposed method is evaluated on the STheReO, FusionPortable, and our self-collected datasets.

This paper presents an improved method for highly accurate identification of the speaker who speaks sentences in the Uzbek language from a limited set of sentences. It is proposed the use of an advanced method for identifying the speaker speaking in the Uzbek language through the use of classification method - Gaussian mixture model (GMM) for decision making. The advanced method is a combination of Gaussian mixture model and MFCC. Current scientific research is aimed at identifying opportunities to improve performance. One of the latest implementations of MFCC are Delta MFCC (DMFCC) and Delta Delta Delta MFCC (DDMFCC), which improve speaker identification. Identification accuracy was obtained from the results of the research. According to the experimental results, speaker identification by conventional MFCC combined with GMM classification algorithm (94.5%) lower than the second (DMFCC+GMM - 96.8%) and third (DDMFCC+GMM-98.1%) feature extraction methods. It is easy to see the performance improvement observed when using DDMFCC in combination with GMM and is 98.1%. first, second, and third level headings (first level heading).

In this work we present a vision-based road user monitoring system for traffic intersections using a combination of Gaussian Mixture Model (GMM)-based deep learning approaches and geometric warping for further behaviour analysis. GMMs and detection of features in consecutive frames are used in bounding box prediction in order to track objects, while a Fast Regional Convolutional Neural Networks (R-CNN) approach classifies different road users. For better analysis, by knowing the actual coordinates of the intersection, we use geometric warping to map the 3D-plane to the 2D plane. We thus extract the real distance and the approximate real time speed of each road user. The results demonstrate that the proposed region proposal generator method for Fast R-CNN outperforms the other methods considered in terms of both accuracy and computation time.

BackgroundAlzheimer's disease (AD) is the most common type of dementia, and early screening is crucial for intervention.ObjectiveCurrently, early screening for older adults without dementia primarily rely on cognitive scale. This study aims to explore a more feasible approach.MethodsPlasma biomarkers (Aβ42/40, p-tau181 and p-tau217) and Gaussian mixture models (GMM) were utilized for stratifying risk levels in older adults without dementia from the Alzheimer's Disease Neuroimaging Initiative. Linear mixed effects model was employed to compare subsequent pathological and cognitive changes, alongside a comparison with traditional scale-based screening methods. Cox regression model was used to assess the risk of progression to dementia across different biomarker status groups.ResultsPlasma Aβ42/40 and p-tau217 effectively predicted Aβ PET pathological progression, while p-tau217 also predicted tau PET changes. All three biomarkers could forecast the progression of FDG PET and cognitive function. P-tau217 and p-tau181 significantly modulated pathology-related cognitive impairment. All three biomarkers could predict dementia risk. The screening method combining GMM with plasma biomarkers demonstrates superior predictive ability compared to traditional scale-based approaches.ConclusionsOur study indicated that the combination of GMM and plasma biomarkers for community screening shows promising potential in monitoring brain health among older adults without dementia. P-tau217 exhibited the best predictive value among the three plasma biomarkers.

A hybrid feature selection scheme and self-organizing map model for machine health assessment

In this paper, we have addressed the problem of real-time detection and tracking of dynamic objects using quadrotors. We have developed a novel object detection algorithm by analyzing and matching the color and spacial features of the target from monocular image sequences. The proposed object detection algorithm can track the objects with high Frame Per Second (FPS) which is suitable for low-end onboard computers that are used in quad-rotors. In addition, we also estimate the position of the target object in real world so that the drone can track the object accurately. A rigorous experimental analysis is provided to show the efficacy of the proposed approach in indoor as well as outdoor environments.

Humans maintain good memory and recognition capability of previous environments when they are learning about new ones. Thus humans are able to continually learn and increase their experience. It is also obvious importance for autonomous mobile robot. The simultaneous localization and mapping system plays an important role in localization and navigation of robot. The loop-closure detection method is an indispensable part of the relocation and map construction, which is critical to correct mappoint errors of simultaneous localization and mapping. Existing visual loop-closure detection methods based on deep learning are not capable of continual learning in terms of cross-scene environment, which bring a great limitation to the application scope. In this article, we propose a novel end-to-end loop-closure detection method based on continual learning, which can effectively suppress the decline of the memory capability of simultaneous localization and mapping system by introducing firstly the orthogonal projection operator into the loop-closure detection to overcome the catastrophic forgetting problem of mobile robot in large-scale and multi-scene environments. Based on the three scenes from public data sets, the experimental results show that the proposed method has a strong capability of continual learning in the cross-scene environment where existing state-of-the-art methods fail.

The loop-closure detection module in visual SLAM eliminates cumulative error from the front-end by detecting the fact that the camera passes through the same place as it passed before, providing the back-end optimization with constraints among longer intervals apart from neighboring frames, resulting in globally consistent trajectories and maps. In order to solve the problem of time-consuming and poor robustness existed in feature point matching based loop-closure detection algorithm, the article uses the Bag-of-words model based on ORB features to find loop-closure and performs spatial consistency verification on results. Firstly, the ORB feature points of image data sets are clustered to construct a dictionary. In order to complete query operations within logarithmic time, the data structure of k-ary tree is adopted. K-means++ clustering is implemented layer by layer to ensure uniform clustering. Afterwards with words bag, an image is expressed by a vector. The article uses Term Frequency-Inverse Document Frequency to assess the importance of each word so as to obtain an more effective description. Finally, the article select key frames to implement the detection. In this way repetitive detection of similar loops can be avoided and coverage of the entire environment can be guaranteed .The L1 norm is used to calculate the similarity scores between key frames. The scores are normalized using a priori similarity thus avoiding the introduction of absolute similarity thresholds, which makes the algorithm more adaptable. To remove wrong loop-closure, the spatial consistency test based on Pose Graph optimization is performed on detection results. Experiments show that the loop-closure detection algorithm used in the article has a good recall rate when the accuracy rate is high.

The aims of this paper are two-fold: to define Gaussian mixture models (GMMs) of colored texture on several feature spaces and to compare the performance of these models in various classification tasks, both with each other and with other models popular in the literature. We construct GMMs over a variety of different color and texture feature spaces, with a view to the retrieval of textured color images from databases. We compare supervised classification results for different choices of color and texture features using the Vistex database, and explore the best set of features and the best GMM configuration for this task. In addition we introduce several methods for combining the ‘color’ and ‘structure’ information in order to improve the classification performances. We then apply the resulting models to the classification of texture databases and to the classification of man-made and natural areas in aerial images. We compare the GMM model with other models in the literature, and show an overall improvement in performance.

We address the problem of localizing objects using color, texture and shape. Given a handrawn sketch for querying an object shape, and its color and texture, the algorithm automatically searches the database images for objects which meet the query attributes. The database images do not need to be presegmented or annotated. The proposed algorithm operates in two stages. In the first stage, we use local texture and color features to find a small number of candidate images, and identify regions in the candidate images which share similar texture and color as the query example. To speed up the processing, the texture and color features are directly extracted from the Discrete Cosine Transform (DCT) compressed domain. In the second stage, we use a deformable template matching method to match the query shape to the image edges at the locations which possess the desired texture and color attributes. This algorithm is different from the other content-based image retrieval algorithms in that: (i) no presegmentation of the database images is needed, and (ii) the color and texture features are directly extracted from the compressed images. Experimental results show that substantial computational savings can be achieved utilizing multiple image cues.

Object localization using color, texture and shape

In this paper, a novel Gaussian mixture fitting filter (GMFF) is proposed to copy with the nonlinear state estimation problem with non-Gaussian measurement environment. The core of GMFF is to use Gaussian mixture regression model to model the unknown measurement likelihood probability, which represents the combination of Gaussian mixture model and linear regression process. In the variational inference framework, through iteratively and alternatively achieving the fitting of the measurement model and the compensation of linear regression error, the estimation accuracy and adaptiveness can be enhanced gradually. The superior performance of GMFF is demonstrated in the simulations.

Fall detection is an important problem in the research of abnormal behavior recognition. In this chapter, a novel real-time method is proposed to detect human fall with a single uncalibrated camera by the changes of global orientation and human shape. Our algorithm has three basic parts: moving object extraction, fall pre-detection, and fall confirmation. The overall orientations are derived from the combination of Gaussian mixture model and motion history image. The shape deformation is quantified from the silhouettes by an approximated bounding box. Standard deviations of the overall orientation and aspect ratio of bounding box are checked to pre-detect a fall, and a fall is confirmed by unmoving shape of the bounding box with the defined fall angle. Experimental results show that our method can detect all possible types of human fall accurately and successfully.KeywordsReal-time Fall DetectionHuman ShapeGlobal OrientationAbnormal Behavior RecognitionMotion History Image (MHI)These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This paper proposes an efficient and robust Loop closure detection (LCD) method based on Convolutional neural network (CNN) feature. The primary method is called SeqCNNSLAM, in which both the outputs of the intermediate layer of a pre-trained CNN and the outputs of traditional sequence-based matching procedure are incorporated, making it possible to handle the viewpoint and condition variance properly. An acceleration algorithm for SeqCNNSLAM is developed to reduce the search range for the current image, resulting in a new LCD method called A-SeqCNNSLAM. To improve the applicability of A-SeqCNNSLAM to new environments, O-SeqCNNSLAM is proposed for online parameters adjustment in A-SeqCNNSLAM. In addition to the above work, we further put forward a promising idea to enhance SeqSLAM by integrating the both CNN features and VLAD's advantages called patch based SeqCNNSLAM (P-SeqCNNSLAM), and provide some preliminary experimental results to reveal its performance.

With the development of automatic driving cars, simultaneous localization and mapping (SLAM) becomes a widely discussed topic. Loop closure detection, as an important module in lidar slam, can provide strong constraint for localization and mapping. In classical lidar SLAM algorithm, traditional loop closure detection method often fails to find out the matching nodes, or even has false positive results, especially in challenging environment like the crossroads or sharp turns. In this paper, a lidar SLAM loop closure detection method using adjacent nodes to check the reliability is introduced. A series of experiments is implemented to verify the effectiveness of the method.