scholarly journals Construction of LWCNN Framework and its Application to Pedestrian Detection with Segmentation Process

2021 ◽  
Vol 3 (3) ◽  
pp. 269-283
Author(s):  
R. Kanthavel
Keyword(s):  
Pedestrian Detection ◽  
Object Identification ◽  
Svm Classifier ◽  
Segmentation Method ◽  
Feature Mapping ◽  
Discussion Section ◽  
Lighting Conditions ◽  
Classification Technique ◽  
Urban Scenario ◽  
Real Traffic

To solve the challenges in traffic object identification, fuzzification, and simplification in a real traffic environment, it is highly required to develop an automatic detection and classification technique for roads, automobiles, and pedestrians with multiple traffic objects inside the same framework. The proposed method has been evaluated on a database with complicated poses, motions, backgrounds, and lighting conditions for an urban scenario where pedestrians are not obstructed. The suggested CNN classifier has an FPR of less than that of the SVM classifier. Confirming the significance of automatically optimized features, the SVM classifier's accuracy is equal to that of the CNN. The proposed framework is integrated with the additional adaptive segmentation method to identify pedestrians more precisely than the conventional techniques. Additionally, the proposed lightweight feature mapping leads to faster calculation times and it has also been verified and tabulated in the results and discussion section.

scholarly journals Bottom-up and Layerwise Domain Adaptation for Pedestrian Detection in Thermal Images

2021 ◽  
Vol 17 (1) ◽  
pp. 1-19
Author(s):  
My Kieu ◽  
Andrew D. Bagdanov ◽  
Marco Bertini
Keyword(s):  
Domain Adaptation ◽  
State Of The Art ◽  
Pedestrian Detection ◽  
Challenging Problem ◽  
Top Down ◽  
Bottom Up ◽  
Security Applications ◽  
Lighting Conditions ◽  
Initial Layers ◽  
Single Modality

Pedestrian detection is a canonical problem for safety and security applications, and it remains a challenging problem due to the highly variable lighting conditions in which pedestrians must be detected. This article investigates several domain adaptation approaches to adapt RGB-trained detectors to the thermal domain. Building on our earlier work on domain adaptation for privacy-preserving pedestrian detection, we conducted an extensive experimental evaluation comparing top-down and bottom-up domain adaptation and also propose two new bottom-up domain adaptation strategies. For top-down domain adaptation, we leverage a detector pre-trained on RGB imagery and efficiently adapt it to perform pedestrian detection in the thermal domain. Our bottom-up domain adaptation approaches include two steps: first, training an adapter segment corresponding to initial layers of the RGB-trained detector adapts to the new input distribution; then, we reconnect the adapter segment to the original RGB-trained detector for final adaptation with a top-down loss. To the best of our knowledge, our bottom-up domain adaptation approaches outperform the best-performing single-modality pedestrian detection results on KAIST and outperform the state of the art on FLIR.

scholarly journals Performance Evaluation of Contour Based Segmentation Methods for Ultrasound Images

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
R. J. Hemalatha ◽  
V. Vijaybaskar ◽  
T. R. Thamizhvani
Keyword(s):  
Active Contour ◽  
Performance Metrics ◽  
Ultrasound Image ◽  
Medical Image Segmentation ◽  
Similarity Criteria ◽  
Support Vector ◽  
Svm Classifier ◽  
Ultrasound Images ◽  
Segmentation Method ◽  
Region Detection

Active contour methods are widely used for medical image segmentation. Using level set algorithms the applications of active contour methods have become flexible and convenient. This paper describes the evaluation of the performance of the active contour models using performance metrics and statistical analysis. We have implemented five different methods for segmenting the synovial region in arthritis affected ultrasound image. A comparative analysis between the methods of segmentation was performed and the best segmentation method was identified using similarity criteria, standard error, and F-test. For further analysis, classification of the segmentation techniques using support vector machine (SVM) classifier is performed to determine the absolute method for synovial region detection. With these results, localized region based active contour named Lankton method is defined to be the best segmentation method.

A rate-of-change-of-current based fault classification technique for thyristor-controlled series-compensated transmission lines

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Nishant H. Kothari ◽  
Bhavesh R. Bhalja ◽  
Vivek Pandya ◽  
Pushkar Tripathi
Keyword(s):  
Transmission Lines ◽  
Noise Signal ◽  
Current Transformer ◽  
Rate Of Change ◽  
Fault Classification ◽  
Support Vector ◽  
Svm Classifier ◽  
Classification Technique ◽  
Suggested Technique ◽  
Zero Sequence

Abstract This paper presents a new fault classification technique for Thyristor-Controlled Series-Compensated (TCSC) transmission lines using Support Vector Machine (SVM). The proposed technique is based on the utilization of post-fault magnitude of Rate-of-Change-of-Current (ROCC). Fault classification has been carried out by giving ROCC of three-phases and zero sequence current as inputs to SVM classifier. The performance of SVM as a binary-class, and multi-class classifier has been evaluated for the proposed feature. The validity of the suggested technique has been tested by modeling a TCSC based 400 kV, 300 km long transmission line using PSCAD/EMTDC software package. Based on the above model, a large number of diversified fault cases (41,220 cases) have been generated by varying fault and system parameters. The effect of window length, current transformer (CT) saturation, noise-signal, and sampling frequency have also been studied. It has been found that the proposed technique provides an accuracy of 99.98% for 37,620 test cases. Moreover, the performance of the suggested technique has also been found to be consistent upon evaluating in a 12-bus power system model consisting of a 365 kV, 60 Hz, 300 km long TCSC line. Comparative evaluation of the proposed SVM based technique with other recent techniques clearly indicates its superiority in terms of fault classification accuracy.

A New Pedestrian Feature Description Method Named Neighborhood Descriptor of Oriented Gradients

2021 ◽  
Vol 16 (1) ◽  
pp. 23-55
Author(s):  
Qian Liu ◽  
Feng Yang ◽  
XiaoFen Tang
Keyword(s):  
Feature Vector ◽  
Pedestrian Detection ◽  
Weight Vector ◽  
Svm Classifier ◽  
Feature Descriptor ◽  
Local Weight ◽  
Detection Approach ◽  
Training Samples ◽  
New Feature ◽  
Linear Svm

In view of the issue of the mechanism for enhancing the neighbourhood relationship of blocks of HOG, this paper proposes neighborhood descriptor of oriented gradients (NDOG), an improved feature descriptor based on HOG, for pedestrian detection. To obtain the NDOG feature vector, the algorithm calculates the local weight vector of the HOG feature descriptor, while integrating spatial correlation among blocks, concatenates this weight vector to the tail of the HOG feature descriptor, and uses the gradient norm to normalize this new feature vector. With the proposed NDOG feature vector along with a linear SVM classifier, this paper develops a complete pedestrian detection approach. Experimental results for the INRIA, Caltech-USA, and ETH pedestrian datasets show that the approach achieves a lower miss rate and a higher average precision compared with HOG and other advanced methods for pedestrian detection especially in the case of insufficient training samples.

Moving Object Classification in a Video Sequence

2017 ◽  
pp. 70-101
Author(s):  
S. Vasavi ◽  
T. Naga Jyothi ◽  
V. Srinivasa Rao
Keyword(s):  
Wavelet Transforms ◽  
Moving Objects ◽  
Performance Test ◽  
Texture Features ◽  
Classification Performance ◽  
Object Identification ◽  
Svm Classifier ◽  
Classification Methods ◽  
Military Installations ◽  
Outdoor Environments

Now-a-day's monitoring objects in a video is a major issue in areas such as airports, banks, military installations. Object identification and recognition are the two important tasks in such areas. These require scanning the entire video which is a time consuming process and hence requires a Robust method to detect and classify the objects. Outdoor environments are more challenging because of occlusion and large distance between camera and moving objects. Existing classification methods have proven to have set of limitations under different conditions. In the proposed system, video is divided into frames and Color features using RGB, HSV histograms, Structure features using HoG, DHoG, Harris, Prewitt, LoG operators and Texture features using LBP, Fourier and Wavelet transforms are extracted. Additionally BoV is used for improving the classification performance. Test results proved that SVM classifier works better compared to Bagging, Boosting, J48 classifiers and works well in outdoor environments.

scholarly journals Research on Fast Pedestrian Detection Algorithm Based on Autoencoding Neural Network and AdaBoost

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-17
Author(s):  
Hongzhi Zhou ◽  
Gan Yu
Keyword(s):  
Neural Network ◽  
Pedestrian Detection ◽  
Detection Algorithm ◽  
Frame Rate ◽  
Original Video ◽  
Video Object ◽  
Sample Weight ◽  
Adaboost Algorithm ◽  
High Level ◽  
Real Traffic

In order to solve the problem of low accuracy of pedestrian detection of real traffic cameras and high missed detection rate of small target pedestrians, this paper combines autoencoding neural network and AdaBoost to construct a fast pedestrian detection algorithm. Aiming at the problem that a single high-level output feature map has insufficient ability to express pedestrian features and existing methods cannot effectively select appropriate multilevel features, this paper improves the traditional AdaBoost algorithm structure, that is, the sample weight update formula and the strong classifier output formula are reset, and the two-input AdaBoost-DBN classification algorithm is proposed. Moreover, in view of the problem that the fusion video is not smoothly played, this paper considers the motion information of the video object, performs pixel interpolation by motion compensation, and restores the frame rate of the original video by reconstructing the dropped interframe image. Through experimental research, we can see that the algorithm constructed in this paper has a certain effect.

Sign in / Sign up

Export Citation Format

Share Document