Abstract:The development of high-performance classifiers represents an important step in improving the timeliness of remote sensing classification in the era of high spatial resolution. Geographically weighted k-nearest neighbors (gwk-NN)-a classifier that incorporates spatial information into the traditional k-NN classifier-has demonstrated to be better at mitigating salt-and-pepper noise and misclassification. However, the integration of spatial dependence relationships into spectral information is computationally intensive. To improve computing performance, this paper discusses two commonly used parallel strategies-data and task parallelism-used to parallelize the gwk-NN classifier in the model training and classification stages, and implements the parallel algorithm by calling MPI and GDAL in the C++ development environment on a standalone eight-core computer. We further investigate the potential performance of dual parallelism (the simultaneous exploitation of data and task parallelism) in image classification. The experimental results demonstrate that the parallel gwk-NN classifier can improve the efficiency of high-resolution, remotely sensed images with multiple land cover types. Specifically, data parallelism is more effective than task parallelism in both model training and classification stages because of the minor role of parallel overhead in total execution time. In addition, dual parallelism can take advantage of data and task parallel strategies in the image classification stage, as evidenced by the two largest speedups attained under dual parallelism I (5.28×) and II (5.73×). Comparatively, dual parallelism II, in which priority is given to data decomposition, achieves the best performance by overlapping computation and data transmission, which is compatible with the current trend toward multicore architectures.

KNN and SVM Parallel algorithm

Topic Detection and Tracking Based on Windowed DBSCAN and Parallel KNN

SVM-KNN Classifier--A New Method of Improving the Accuracy of SVM Classifier

Research on Parallelization of GPU-based K-Nearest Neighbor Algorithm

A NEP-SVM based simulation system for event tracking

A Parallel Computing Algorithm for Moving Targets Tracking in Wireless Sensor Networks

Design and implementation of a parallel geographically weighted k-nearest neighbor classifier

Efficient Parallel Processing of High-Dimensional Spatial K NN Queries

A System for Event Detection and Tracking Based on Constructive-Competition Clustering and KNNFL

The Research of the Parallel Smo Algorithm for Solving Svm

Optimization of Multi Kernel Parallel Support Vector Machine Based on Hadoop

Design and implementation of a parallel geographically weighted -nearest neighbor classifier.

PEDESTRIAN DETECTION AND TRACKING OF THEIR MOVEMENT TRAJECTORY USING THE BACKGROUND SEGMENTATION METHOD BASED ON KNN

Efficient parallel algorithm for linear RankSVM on multi-core systems

Parallel multi-label K-nearest neighbor algorithm based on Spark

Research on Topic Detection and Tracking for Online News Texts

Parallel Implementation of Gradient-Based Neural Networks for Svm Training

Parallel Search Strategy in Kernel Feature Space to Track FLIR Target

Application of SVM-KNN Combination Improvement Algorithm on Patent Text Classification

A New Parallel Video Understanding and Retrieval System

Parallel and memory-efficient realization of KSP/KNN algorithm