Dataset Information

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering.

ABSTRACT: The object detection algorithm based on vehicle-mounted lidar is a key component of the perception system on autonomous vehicles. It can provide high-precision and highly robust obstacle information for the safe driving of autonomous vehicles. However, most algorithms are often based on a large amount of point cloud data, which makes real-time detection difficult. To solve this problem, this paper proposes a 3D fast object detection method based on three main steps: First, the ground segmentation by discriminant image (GSDI) method is used to convert point cloud data into discriminant images for ground points segmentation, which avoids the direct computing of the point cloud data and improves the efficiency of ground points segmentation. Second, the image detector is used to generate the region of interest of the three-dimensional object, which effectively narrows the search range. Finally, the dynamic distance threshold clustering (DDTC) method is designed for different density of the point cloud data, which improves the detection effect of long-distance objects and avoids the over-segmentation phenomenon generated by the traditional algorithm. Experiments have showed that this algorithm can meet the real-time requirements of autonomous driving while maintaining high accuracy.

SUBMITTER: Chen B

PROVIDER: S-EPMC7766091 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering.

Chen Baifan B Chen Hong H Yuan Dian D Yu Lingli L

Sensors (Basel, Switzerland) 20201217 24

The object detection algorithm based on vehicle-mounted lidar is a key component of the perception system on autonomous vehicles. It can provide high-precision and highly robust obstacle information for the safe driving of autonomous vehicles. However, most algorithms are often based on a large amount of point cloud data, which makes real-time detection difficult. To solve this problem, this paper proposes a 3D fast object detection method based on three main steps: First, the ground segmentatio ...[more]

PMID: 33348559

Similar Datasets

Project description:The benefits of integrating attention and object recognition are investigated. While attention is frequently modeled as a pre-processor for recognition, we investigate the hypothesis that attention is an intrinsic component of recognition and vice-versa. This hypothesis is tested with a recognition model, the hierarchical discriminant saliency network (HDSN), whose layers are top-down saliency detectors, tuned for a visual class according to the principles of discriminant saliency. As a model of neural computation, the HDSN has two possible implementations. In a biologically plausible implementation, all layers comply with the standard neurophysiological model of visual cortex, with sub-layers of simple and complex units that implement a combination of filtering, divisive normalization, pooling, and non-linearities. In a convolutional neural network implementation, all layers are convolutional and implement a combination of filtering, rectification, and pooling. The rectification is performed with a parametric extension of the now popular rectified linear units (ReLUs), whose parameters can be tuned for the detection of target object classes. This enables a number of functional enhancements over neural network models that lack a connection to saliency, including optimal feature denoising mechanisms for recognition, modulation of saliency responses by the discriminant power of the underlying features, and the ability to detect both feature presence and absence. In either implementation, each layer has a precise statistical interpretation, and all parameters are tuned by statistical learning. Each saliency detection layer learns more discriminant saliency templates than its predecessors and higher layers have larger pooling fields. This enables the HDSN to simultaneously achieve high selectivity to target object classes and invariance. The performance of the network in saliency and object recognition tasks is compared to those of models from the biological and computer vision literatures. This demonstrates benefits for all the functional enhancements of the HDSN, the class tuning inherent to discriminant saliency, and saliency layers based on templates of increasing target selectivity and invariance. Altogether, these experiments suggest that there are non-trivial benefits in integrating attention and recognition.

Project description:Representing the 3D structures of ligands in virtual screenings via multi-conformer ensembles can be computationally intensive, especially for compounds with a large number of rotatable bonds. Thus, reducing the size of multi-conformer databases and the number of query conformers, while simultaneously reproducing the bioactive conformer with good accuracy, is of crucial interest. While clustering and RMSD filtering methods are employed in existing conformer generators, the novelty of this work is the inclusion of a clustering scheme (NMRCLUST) that does not require a user-defined cut-off value. This algorithm simultaneously optimizes the number and the average spread of the clusters. Here we describe and test four inter-dependent approaches for selecting computer-generated conformers, namely: OMEGA, NMRCLUST, RMS filtering and averaged-RMS filtering. The bioactive conformations of 65 selected ligands were extracted from the corresponding protein:ligand complexes from the Protein Data Bank, including eight ligands that adopted dissimilar bound conformations within different receptors. We show that NMRCLUST can be employed to further filter OMEGA-generated conformers while maintaining biological relevance of the ensemble. It was observed that NMRCLUST (containing on average 10 times fewer conformers per compound) performed nearly as well as OMEGA, and both outperformed RMS filtering and averaged-RMS filtering in terms of identifying the bioactive conformations with excellent and good matches (0.5 < RMSD < 1.0 A). Furthermore, we propose thresholds for OMEGA root-mean square filtering depending on the number of rotors in a compound: 0.8, 1.0 and 1.4 for structures with low (1-4), medium (5-9) and high (10-15) numbers of rotatable bonds, respectively. The protocol employed is general and can be applied to reduce the number of conformers in multi-conformer compound collections and alleviate the complexity of downstream data processing in virtual screening experiments.

Dataset Information

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering.

Publications

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets