scholarly journals A Shallow–Deep Feature Fusion Method for Pedestrian Detection

2021 ◽  
Vol 11 (19) ◽  
pp. 9202
Author(s):  
Daxue Liu ◽  
Kai Zang ◽  
Jifeng Shen
Keyword(s):  
Detection Method ◽  
Feature Fusion ◽  
Pedestrian Detection ◽  
Recall Rate ◽  
Fusion Method ◽  
Discriminative Power ◽  
Deep Feature ◽  
Proposal A ◽  
High Recall Rate ◽  
Feature Based

In this paper, a shallow–deep feature fusion (SDFF) method is developed for pedestrian detection. Firstly, we propose a shallow feature-based method under the ACF framework of pedestrian detection. More precisely, improved Haar-like templates with Local FDA learning are used to filter the channel maps of ACF such that these Haar-like features are able to improve the discriminative power and therefore enhance the detection performance. The proposed shallow feature is also referred to as weighted subset-haar-like feature. It is efficient in pedestrian detection with a high recall rate and precise localization. Secondly, the proposed shallow feature-based detection method operates as a region proposal. A classifier equipped with ResNet is then used to refine the region proposals to judge whether each region contains a pedestrian or not. The extensive experiments evaluated on INRIA, Caltech, and TUD-Brussel datasets show that SDFF is an effective and efficient method for pedestrian detection.

scholarly journals Small-scale Pedestrian Detection Method based on Attention Mechanism and Adaptive Feature Fusion

2021 ◽  
Vol 2035 (1) ◽  
pp. 012023
Author(s):  
Yuhao You ◽  
Houjin Chen ◽  
Yanfeng Li ◽  
Minjun Wang ◽  
Jinlei Zhu
Keyword(s):  
Detection Method ◽  
Feature Fusion ◽  
Pedestrian Detection ◽  
Attention Mechanism ◽  
Small Scale

scholarly journals Remote Sensing and Texture Image Classification Network Based on Deep Learning Integrated with Binary Coding and Sinkhorn Distance

2019 ◽  
Vol 11 (23) ◽  
pp. 2870
Author(s):  
Chu He ◽  
Qingyi Zhang ◽  
Tao Qu ◽  
Dingwen Wang ◽  
Mingsheng Liao
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
Image Classification ◽  
Feature Fusion ◽  
Texture Features ◽  
Texture Image ◽  
Binary Coding ◽  
Deep Feature ◽  
Feature Based ◽  
Texture Image Classification

In the past two decades, traditional hand-crafted feature based methods and deep feature based methods have successively played the most important role in image classification. In some cases, hand-crafted features still provide better performance than deep features. This paper proposes an innovative network based on deep learning integrated with binary coding and Sinkhorn distance (DBSNet) for remote sensing and texture image classification. The statistical texture features of the image extracted by uniform local binary pattern (ULBP) are introduced as a supplement for deep features extracted by ResNet-50 to enhance the discriminability of features. After the feature fusion, both diversity and redundancy of the features have increased, thus we propose the Sinkhorn loss where an entropy regularization term plays a key role in removing redundant information and training the model quickly and efficiently. Image classification experiments are performed on two texture datasets and five remote sensing datasets. The results show that the statistical texture features of the image extracted by ULBP complement the deep features, and the new Sinkhorn loss performs better than the commonly used softmax loss. The performance of the proposed algorithm DBSNet ranks in the top three on the remote sensing datasets compared with other state-of-the-art algorithms.

scholarly journals Recognition of Fetal Facial Ultrasound Standard Plane Based on Texture Feature Fusion

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Xiaoli Wang ◽  
Zhonghua Liu ◽  
Yongzhao Du ◽  
Yong Diao ◽  
Peizhong Liu ◽  
...  
Keyword(s):  
Cleft Lip ◽  
Cleft Lip And Palate ◽  
Feature Fusion ◽  
Texture Feature ◽  
Recall Rate ◽  
Support Vector ◽  
Ultrasound Images ◽  
Fusion Method ◽  
Accurate Identification ◽  
Data Set

In the process of prenatal ultrasound diagnosis, accurate identification of fetal facial ultrasound standard plane (FFUSP) is essential for accurate facial deformity detection and disease screening, such as cleft lip and palate detection and Down syndrome screening check. However, the traditional method of obtaining standard planes is manual screening by doctors. Due to different levels of doctors, this method often leads to large errors in the results. Therefore, in this study, we propose a texture feature fusion method (LH-SVM) for automatic recognition and classification of FFUSP. First, extract image’s texture features, including Local Binary Pattern (LBP) and Histogram of Oriented Gradient (HOG), then perform feature fusion, and finally adopt Support Vector Machine (SVM) for predictive classification. In our study, we used fetal facial ultrasound images from 20 to 24 weeks of gestation as experimental data for a total of 943 standard plane images (221 ocular axial planes, 298 median sagittal planes, 424 nasolabial coronal planes, and 350 nonstandard planes, OAP, MSP, NCP, N-SP). Based on this data set, we performed five-fold cross-validation. The final test results show that the accuracy rate of the proposed method for FFUSP classification is 94.67%, the average precision rate is 94.27%, the average recall rate is 93.88%, and the average F 1 score is 94.08%. The experimental results indicate that the texture feature fusion method can effectively predict and classify FFUSP, which provides an essential basis for clinical research on the automatic detection method of FFUSP.

scholarly journals Analysis of Feature Fusion Based on HIK SVM and Its Application for Pedestrian Detection

2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
Song-Zhi Su ◽  
Shu-Yuan Chen
Keyword(s):  
Near Infrared ◽  
Feature Fusion ◽  
Pedestrian Detection ◽  
Visual Images ◽  
Support Vector ◽  
Fusion Method ◽  
Low Resolution ◽  
Night Time ◽  
Histogram Of Oriented Gradient ◽  
Histogram Intersection

This work presents the fusion of integral channel features to improve the effectiveness and efficiency of pedestrian detection. The proposed method combines the histogram of oriented gradient (HOG) and local binary pattern (LBP) features by a concatenated fusion method. Although neural network (NN) is an efficient tool for classification, the time complexity is heavy. Hence, we choose support vector machine (SVM) with the histogram intersection kernel (HIK) as a classifier. On the other hand, although many datasets have been collected for pedestrian detection, few are designed to detect pedestrians in low-resolution visual images and at night time. This work collects two new pedestrian datasets—one for low-resolution visual images and one for near-infrared images—to evaluate detection performance on various image types and at different times. The proposed fusion method uses only images from the INRIA dataset for training but works on the two newly collected datasets, thereby avoiding the training overhead for cross-datasets. The experimental results verify that the proposed method has high detection accuracies even in the variations of image types and time slots.

scholarly journals Deep Feature Fusion by Competitive Attention for Pedestrian Detection

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 21981-21989 ◽  
Author(s):  
Zhichang Chen ◽  
Li Zhang ◽  
Abdul Mateen Khattak ◽  
Wanlin Gao ◽  
Minjuan Wang
Keyword(s):  
Feature Fusion ◽  
Pedestrian Detection ◽  
Deep Feature

A Multi-Feature Fusion Method for Forward Vehicle Detection with Single Camera

2013 ◽  
Vol 321-324 ◽  
pp. 998-1004 ◽  
Author(s):  
Xing Li ◽  
Xiao Song Guo ◽  
Jun Bin Guo
Keyword(s):  
Shape Matching ◽  
Feature Fusion ◽  
Detection System ◽  
Vehicle Detection ◽  
Single Camera ◽  
Fusion Method ◽  
Threshold Method ◽  
Feature Based ◽  
Matching Degree ◽  
Edge Features

Vehicle detection is very important for Advanced Driver Assistance System. This paper focused on improving the performance of vehicle detection system with single camera and proposed a multi-feature fusion method for forward vehicle detection. The shadow and edges of the vehicle are the most important features, so they can be utilized to detect vehicle at daytime. The shadow and edge features were segmented accurately by using histogram analysis method and adaptive dual-threshold method respectively. The initial candidates were generated by combining edge and shadow features, and these initial candidates were further verified using an integrated feature based on the fusion of symmetry, texture and shape matching degree features. The weight of each feature was determined by the Fisher criterion, and the non-vehicle initial candidates were rejected by a threshold. The experimental results show that the proposed method could be adapt to different illumination circumstances robustly and improve the accuracy of forward vehicle detection.

Sign in / Sign up

Export Citation Format

Share Document