Financial Topic Detection Algorithm Based on Multi-feature Fusion

2020 ◽  
pp. 11-19
Author(s):  
Qiang Zhang ◽  
Junping Du ◽  
Feifei Kou
Keyword(s):  
Feature Fusion ◽  
Detection Algorithm ◽  
Topic Detection

scholarly journals Pedestrian detection algorithm based on improved muti-scale feature fusion

2021 ◽  
Vol 2078 (1) ◽  
pp. 012008
Author(s):  
Hui Liu ◽  
Keyang Cheng
Keyword(s):  
Clustering Algorithm ◽  
Feature Fusion ◽  
Pedestrian Detection ◽  
Detection Algorithm ◽  
Data Sets ◽  
False Detection ◽  
Scale Feature ◽  
Multi Scale ◽  
Dilated Convolution ◽  
Small Targets

Abstract Aiming at the problem of false detection and missed detection of small targets and occluded targets in the process of pedestrian detection, a pedestrian detection algorithm based on improved multi-scale feature fusion is proposed. First, for the YOLOv4 multi-scale feature fusion module PANet, which does not consider the interaction relationship between scales, PANet is improved to reduce the semantic gap between scales, and the attention mechanism is introduced to learn the importance of different layers to strengthen feature fusion; then, dilated convolution is introduced. Dilated convolution reduces the problem of information loss during the downsampling process; finally, the K-means clustering algorithm is used to redesign the anchor box and modify the loss function to detect a single category. The experimental results show that the improved pedestrian detection algorithm in the INRIA and WiderPerson data sets under different congestion conditions, the AP reaches 96.83% and 59.67%, respectively. Compared with the pedestrian detection results of the YOLOv4 model, the algorithm improves by 2.41% and 1.03%, respectively. The problem of false detection and missed detection of small targets and occlusion has been significantly improved.

Detection and Tracking Cows by Computer Vision and Image Classification Methods

2021 ◽  
Vol 13 (1) ◽  
pp. 1-45
Author(s):  
Terry Gao
Keyword(s):  
Feature Fusion ◽  
Binary Classification ◽  
Classification Problem ◽  
Detection Algorithm ◽  
Video Sequences ◽  
Image Block ◽  
Body Contour ◽  
Adaboost Algorithm ◽  
Target Model ◽  
Detection And Tracking

In this paper, the cow recognition and traction in video sequences is studied. In the recognition phase, this paper does some discussion and analysis which aim at different classification algorithms and feature extraction algorithms, and cow's detection is transformed into a binary classification problem. The detection method extracts cow's features using a method of multiple feature fusion. These features include edge characters which reflects the cow body contour, grey value, and spatial position relationship. In addition, the algorithm detects the cow body through the classifier which is trained by Gentle Adaboost algorithm. Experiments show that the method has good detection performance when the target has deformation or the contrast between target and background is low. Compared with the general target detection algorithm, this method reduces the miss rate and the detection precision is improved. Detection rate can reach 97.3%. In traction phase, the popular compressive tracking (CT) algorithm is proposed. The learning rate is changed through adaptively calculating the pap distance of image block. Moreover, the update for target model is stopped to avoid introducing error and noise when the classification response values are negative. The experiment results show that the improved tracking algorithm can effectively solve the target model update by mistaken when there are large covers or the attitude is changed frequently. For the detection and tracking of cow body, a detection and tracking framework for the image of cow is built and the detector is combined with the tracking framework. The algorithm test for some video sequences under the complex environment indicates the detection algorithm based on improved compressed perception shows good tracking effect in the changing and complicated background.

scholarly journals DF-SSD: An Improved SSD Object Detection Algorithm Based on DenseNet and Feature Fusion

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 24344-24357 ◽  
Author(s):  
Sheping Zhai ◽  
Dingrong Shang ◽  
Shuhuan Wang ◽  
Susu Dong
Keyword(s):  
Object Detection ◽  
Feature Fusion ◽  
Detection Algorithm

Multi-level topic detection algorithm for Netnews Specials

2013 ◽  
Author(s):  
Yu Peng ◽  
ZhiQing Lin ◽  
Bo Xiao ◽  
Chuang Zhang
Keyword(s):  
Detection Algorithm ◽  
Topic Detection ◽  
Multi Level

scholarly journals Shot Boundary Detection Algorithm Based on Multi-Feature Fusion

2013 ◽  
Vol 347-350 ◽  
pp. 3866-3871
Author(s):  
Kai Jin ◽  
Hong Cai Feng ◽  
Qi Feng ◽  
Chi Zhang
Keyword(s):  
Feature Fusion ◽  
Texture Feature ◽  
Boundary Detection ◽  
Detection Algorithm ◽  
Shot Boundary Detection ◽  
Features Fusion ◽  
Orientation Histogram ◽  
Video Shot ◽  
Shot Boundary ◽  
The Ideal

To establish a general and robust shot boundary detection algorithm, according to characteristics of lens conversion and the ideal of multiple video features fusion, a shot boundary detection algorithm is proposed based on YUV histogram, texture feature and edge orientation histogram in the paper. Besides, global and self-adaptive threshold are combined to use so as to control the process of shot boundary detection and enhance the accuracy of threshold selection. The experiment results show that the algorithm can effectively realize video shot boundary detection and strengthen the robustness of the detection.

Studies on a multidimensional public opinion network model and its topic detection algorithm

2019 ◽  
Vol 56 (3) ◽  
pp. 584-608 ◽  
Author(s):  
Guanghui Wang ◽  
Yuxue Chi ◽  
Yijun Liu ◽  
Yufei Wang
Keyword(s):  
Public Opinion ◽  
Network Model ◽  
Detection Algorithm ◽  
Topic Detection

The Blog Topic Detection Research Based on Bursty Word

2014 ◽  
Vol 926-930 ◽  
pp. 3406-3409
Author(s):  
Tao Kuang ◽  
Shan Hong Zhu
Keyword(s):  
Similarity Measure ◽  
Clustering Analysis ◽  
Main Idea ◽  
Design Strategy ◽  
Detection Algorithm ◽  
Experimental Results ◽  
Topic Detection ◽  
Media Report ◽  
Word Similarity ◽  
Detecting Method

The emergence of blog hot topic means that the user's interest ,participation behavior and various media report coverage reach to its climax,a detecting method of topics on blog based on blog bursty words is proposed. It includes the use of word similarity measure and text clustering analysis which is combined with design strategy in specific period, the use of the main idea of the sudden vocabulary hot topic detection algorithm has to be used and improved in order to generate the final clustering. The experimental results show that the algorithm can obtain an accurate blog topic detection results.

Sign in / Sign up

Export Citation Format

Share Document