Generic Object Detection Based on Boosting Embedded with Bag-of-Words

2011 ◽  
Vol 109 ◽  
pp. 285-289
Author(s):  
Xue Na Qiu ◽  
Shi Rong Liu ◽  
Jia Tao Song
Keyword(s):  
Object Detection ◽  
Detection Efficiency ◽  
Classification Performance ◽  
Complex Object ◽  
Detection Accuracy ◽  
Bag Of Words ◽  
Good Classification ◽  
Local Patch ◽  
Good Classification Performance ◽  
Boosting Method

This paper studies generic object detection. In the view of complexity and diversity of generic object, it proposes Boosting generic object detection method with bag-of-words. Boosting method has good detection efficiency, but it has some fault detections due to the diversity and complexity of the object. While Bag-of-words method has some advantages, such as local patch features, simplicity and robustness, and it has good classification performance of complex object. The proposed method applies Bag-of-words to remove the fault detection and to improve the tracking results of Boosting, and thus it achieves high generic object detection accuracy.

Defect Classification in Pipes by Neural Networks Using Multiple Guided Ultrasonic Wave Features Extracted After Wavelet Processing

2005 ◽  
Vol 127 (3) ◽  
pp. 294-303 ◽  
Author(s):  
Piervincenzo Rizzo ◽  
Ivan Bartoli ◽  
Alessandro Marzani ◽  
Francesco Lanza di Scalea
Keyword(s):  
Guided Waves ◽  
Fourier Transforms ◽  
Damage Index ◽  
Classification Performance ◽  
Discrete Wavelet ◽  
Cross Sectional ◽  
Classification Framework ◽  
Discrete Wavelet Decomposition ◽  
Good Classification ◽  
Good Classification Performance

This paper casts pipe inspection by ultrasonic guided waves in a feature extraction and automatic classification framework. The specific defect under investigation is a small notch cut in an ASTM-A53-F steel pipe at depths ranging from 1% to 17% of the pipe cross-sectional area. A semi-analytical finite element method is first used to model wave propagation in the pipe. In the experiment, reflection measurements are taken and six features are extracted from the discrete wavelet decomposition of the raw signals and from the Hilbert and Fourier transforms of the reconstructed signals. A six-dimensional damage index is then constructed, and it is fed to an artificial neural network that classifies the size and the location of the notch. Overall, the wavelet-based multifeature analysis demonstrates good classification performance and robustness against noise and changes in some of the operating parameters.

IMPROVING SVM PERFORMANCE IN MULTI-LABEL DOMAINS: THRESHOLD ADJUSTMENT

2013 ◽  
Vol 22 (01) ◽  
pp. 1250038 ◽  
Author(s):  
PEERAPON VATEEKUL ◽  
SAREEWAN DENDAMRONGVIT ◽  
MIROSLAV KUBAT
Keyword(s):  
Classification Performance ◽  
Support Vector ◽  
Binary Classifier ◽  
Research Groups ◽  
Computational Costs ◽  
Vector Machines ◽  
Threshold Adjustment ◽  
Good Classification ◽  
Good Classification Performance ◽  
Suboptimal Behavior

In “multi-label domains,” where the same example can simultaneously belong to two or more classes, it is customary to induce a separate binary classifier for each class, and then use them all in parallel. As a result, some of these classifiers are induced from imbalanced training sets where one class outnumbers the other – a circumstance known to hurt some machine learning paradigms. In the case of Support Vector Machines (SVM), this suboptimal behavior is explained by the fact that SVM seeks to minimize error rate, a criterion that is in domains of this type misleading. This is why several research groups have studied mechanisms to readjust the bias of SVM's hyperplane. The best of these achieves very good classification performance at the price of impractically high computational costs. We propose here an improvement where these cost are reduced to a small fraction without significantly impairing classification.

scholarly journals A New Edge Patch with Rotation Invariance for Object Detection and Pose Estimation

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 887 ◽  
Author(s):  
Xunwei Tong ◽  
Ruifeng Li ◽  
Lianzheng Ge ◽  
Lijun Zhao ◽  
Ke Wang
Keyword(s):  
Object Detection ◽  
Pose Estimation ◽  
Rotation Invariance ◽  
Depth Image ◽  
Detection Accuracy ◽  
Edge Pixel ◽  
Local Patch ◽  
Background Clutter ◽  
Public Datasets ◽  
Local Sampling

Local patch-based methods of object detection and pose estimation are promising. However, to the best of the authors’ knowledge, traditional red-green-blue and depth (RGB-D) patches contain scene interference (foreground occlusion and background clutter) and have little rotation invariance. To solve these problems, a new edge patch is proposed and experimented with in this study. The edge patch is a local sampling RGB-D patch centered at the edge pixel of the depth image. According to the normal direction of the depth edge, the edge patch is sampled along a canonical orientation, making it rotation invariant. Through a process of depth detection, scene interference is eliminated from the edge patch, which improves the robustness. The framework of the edge patch-based method is described, and the method was evaluated on three public datasets. Compared with existing methods, the proposed method achieved a higher average F1-score (0.956) on the Tejani dataset and a better average detection rate (62%) on the Occlusion dataset, even in situations of serious scene interference. These results showed that the proposed method has higher detection accuracy and stronger robustness.

scholarly journals Training Classifiers under Covariate Shift by Constructing the Maximum Consistent Distribution Subset

2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Xu Yu ◽  
Miao Yu ◽  
Li-xun Xu ◽  
Jing Yang ◽  
Zhi-qiang Xie
Keyword(s):  
Subset Selection ◽  
Feature Space ◽  
Classification Performance ◽  
Selection Method ◽  
Experimental Results ◽  
Training Set ◽  
Traditional Methods ◽  
Generalization Ability ◽  
Good Classification ◽  
Good Classification Performance

The assumption that the training and testing samples are drawn from the same distribution is violated under covariate shift setting, and most algorithms for the covariate shift setting try to first estimate distributions and then reweight samples based on the distributions estimated. Due to the difficulty of estimating a correct distribution, previous methods can not get good classification performance. In this paper, we firstly present two types of covariate shift problems. Rather than estimating the distributions, we then desire an effective method to select a maximum subset following the target testing distribution based on feature space split from the auxiliary set or the target training set. Finally, we prove that our subset selection method can consistently deal with both scenarios of covariate shift. Experimental results demonstrate that training a classifier with the selected maximum subset exhibits good generalization ability and running efficiency over those of traditional methods under covariate shift setting.

scholarly journals Coupled Least Squares Support Vector Ensemble Machines

Information ◽  
2019 ◽  
Vol 10 (6) ◽  
pp. 195 ◽  
Author(s):  
Dickson Keddy Wornyo ◽  
Xiang-Jun Shen
Keyword(s):  
Least Squares ◽  
Classification Performance ◽  
Support Vector ◽  
Vector Method ◽  
Wide Range ◽  
Good Classification ◽  
Coupling Strategy ◽  
Good Classification Performance ◽  
Data Driven Modeling ◽  
Model Approach

The least squares support vector method is a popular data-driven modeling method which shows better performance and has been successfully applied in a wide range of applications. In this paper, we propose a novel coupled least squares support vector ensemble machine (C-LSSVEM). The proposed coupling ensemble helps improve robustness and produce good classification performance than the single model approach. The proposed C-LSSVEM can choose appropriate kernel types and their parameters in a good coupling strategy with a set of classifiers being trained simultaneously. The proposed method can further minimize the total loss of ensembles in kernel space. Thus, we form an ensemble regressor by co-optimizing and weighing base regressors. Experiments conducted on several datasets such as artificial datasets, UCI classification datasets, UCI regression datasets, handwritten digits datasets and NWPU-RESISC45 datasets, indicate that C-LSSVEM performs better in achieving the minimal regression loss and the best classification accuracy relative to selected state-of-the-art regression and classification techniques.

Application of WAVELET-SVM in Fault Diagnosis for the UV Control System

2011 ◽  
Vol 65 ◽  
pp. 199-203
Author(s):  
Sheng Wu Wang ◽  
Xiu Hua Shi ◽  
Hui Xu ◽  
Zhao Jing Tong
Keyword(s):  
Control System ◽  
Fault Diagnosis ◽  
Wavelet Transformation ◽  
Classification Performance ◽  
Support Vector ◽  
Fault Features ◽  
Good Classification ◽  
Good Classification Performance ◽  
Simulation Results ◽  
Fault Characteristic

Wavelet Analysis extracts the main feature from the fault signal through wavelet transformation, so it is advantageous to withdraw fault characteristic for fault diagnosis. Support Vector Machine (SVM) has shown its good classification performance in fault diagnosis. A new method of fault diagnosis for UV control system based on WAVELET-SVM is raised. The sensor output is sampled in frequency domain and it is preprocessed by wavelet to extract main vectors of the fault features. Fault patterns under various states are classified using multi-class SVM, and fault diagnosis is realized. The simulation results show that WAVELET-SVM is feasible to detect and locate faults quickly and exactly and has high robustness.

Combined RF-based drone detection and classification

2021 ◽  
Author(s):  
Sanjoy Basak ◽  
Sreeraj Rajendran ◽  
Sofie Pollin ◽  
Bart Scheers
Keyword(s):  
Nuclear Power ◽  
Power Plants ◽  
Goodness Of Fit ◽  
Detailed Comparison ◽  
Classification Performance ◽  
Classification Framework ◽  
Good Classification ◽  
Good Classification Performance ◽  
Transmitted Signal ◽  
Combined Detection

Despite several beneficial applications, unfortunately, drones are also being used for illicit activities such as drug trafficking, firearm smuggling or to impose threats to security-sensitive places like airports and nuclear power plants. The existing drone localization and neutralization technologies work on the assumption that the drone has already been detected and classified. Although we have observed a tremendous advancement in the sensor industry in this decade, there is no robust drone detection and classification method proposed in the literature yet. This paper focuses on radio frequency (RF) based drone detection and classification using the frequency signature of the transmitted signal. We have created a novel drone RF dataset using commercial drones and presented a detailed comparison between a two-stage and combined detection and classification framework. The detection and classification performance of both frameworks are presented for a single-signal and simultaneous multi-signal scenario. With detailed analysis, we show that You Only Look Once (YOLO) framework provides better detection performance compared to the Goodness-of-Fit (GoF) spectrum sensing for a simultaneous multi-signal scenario and good classification performance comparable to Deep Residual Neural Network (DRNN) framework.<br>

scholarly journals Combined RF-based drone detection and classification

2021 ◽  
Author(s):  
Sanjoy Basak ◽  
Sreeraj Rajendran ◽  
Sofie Pollin ◽  
Bart Scheers
Keyword(s):  
Nuclear Power ◽  
Power Plants ◽  
Goodness Of Fit ◽  
Detailed Comparison ◽  
Classification Performance ◽  
Classification Framework ◽  
Good Classification ◽  
Good Classification Performance ◽  
Transmitted Signal ◽  
Combined Detection

Despite several beneficial applications, unfortunately, drones are also being used for illicit activities such as drug trafficking, firearm smuggling or to impose threats to security-sensitive places like airports and nuclear power plants. The existing drone localization and neutralization technologies work on the assumption that the drone has already been detected and classified. Although we have observed a tremendous advancement in the sensor industry in this decade, there is no robust drone detection and classification method proposed in the literature yet. This paper focuses on radio frequency (RF) based drone detection and classification using the frequency signature of the transmitted signal. We have created a novel drone RF dataset using commercial drones and presented a detailed comparison between a two-stage and combined detection and classification framework. The detection and classification performance of both frameworks are presented for a single-signal and simultaneous multi-signal scenario. With detailed analysis, we show that You Only Look Once (YOLO) framework provides better detection performance compared to the Goodness-of-Fit (GoF) spectrum sensing for a simultaneous multi-signal scenario and good classification performance comparable to Deep Residual Neural Network (DRNN) framework.<br>

scholarly journals A Study on Enhancement of Fish Recognition Using Cumulative Mean of YOLO Network in Underwater Video Images

2020 ◽  
Vol 8 (11) ◽  
pp. 952
Author(s):  
Jin-Hyun Park ◽  
Changgu Kang
Keyword(s):  
Classification Performance ◽  
Current Frame ◽  
Exotic Invasive Species ◽  
Video Images ◽  
Underwater Environment ◽  
Good Classification ◽  
Exotic Invasive ◽  
Rare And Endangered ◽  
Good Classification Performance ◽  
Classification Probability

In the underwater environment, in order to preserve rare and endangered objects or to eliminate the exotic invasive species that can destroy the ecosystems, it is essential to classify objects and estimate their number. It is very difficult to classify objects and estimate their number. While YOLO shows excellent performance in object recognition, it recognizes objects by processing the images of each frame independently of each other. By accumulating the object classification results from the past frames to the current frame, we propose a method to accurately classify objects, and count their number in sequential video images. This has a high classification probability of 93.94% and 97.06% in the test videos of Bluegill and Largemouth bass, respectively. The proposed method shows very good classification performance in video images taken of the underwater environment.

scholarly journals A Slimmer Network with Polymorphic and Group Attention Modules for More Efficient Object Detection in Aerial Images

2020 ◽  
Vol 12 (22) ◽  
pp. 3750
Author(s):  
Wei Guo ◽  
Weihong Li ◽  
Zhenghao Li ◽  
Weiguo Gong ◽  
Jinkai Cui ◽  
...  
Keyword(s):  
Object Detection ◽  
Detection Efficiency ◽  
Aerial Images ◽  
Aerial Image ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Practical Applications ◽  
Multi Scale ◽  
High Detection Efficiency ◽  
Object Features

Object detection is one of the core technologies in aerial image processing and analysis. Although existing aerial image object detection methods based on deep learning have made some progress, there are still some problems remained: (1) Most existing methods fail to simultaneously consider multi-scale and multi-shape object characteristics in aerial images, which may lead to some missing or false detections; (2) high precision detection generally requires a large and complex network structure, which usually makes it difficult to achieve the high detection efficiency and deploy the network on resource-constrained devices for practical applications. To solve these problems, we propose a slimmer network for more efficient object detection in aerial images. Firstly, we design a polymorphic module (PM) for simultaneously learning the multi-scale and multi-shape object features, so as to better detect the hugely different objects in aerial images. Then, we design a group attention module (GAM) for better utilizing the diversiform concatenation features in the network. By designing multiple detection headers with adaptive anchors and the above-mentioned two modules, we propose a one-stage network called PG-YOLO for realizing the higher detection accuracy. Based on the proposed network, we further propose a more efficient channel pruning method, which can slim the network parameters from 63.7 million (M) to 3.3M that decreases the parameter size by 94.8%, so it can significantly improve the detection efficiency for real-time detection. Finally, we execute the comparative experiments on three public aerial datasets, and the experimental results show that the proposed method outperforms the state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document