scholarly journals Pedestrian detection algorithm in traffic scene based on weakly supervised hierarchical deep model

2016 ◽  
Vol 14 (1) ◽  
pp. 172988141769231 ◽  
Author(s):  
Yingfeng Cai ◽  
Youguo He ◽  
Hai Wang ◽  
Xiaoqiang Sun ◽  
Long Chen ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pedestrian Detection ◽  
Recognition Rate ◽  
Detection Algorithm ◽  
Training Methods ◽  
Two Dimensional ◽  
Data Set ◽  
Deep Model ◽  
Unsupervised Training ◽  
Weakly Supervised

The emergence and development of deep learning theory in machine learning field provide new method for visual-based pedestrian recognition technology. To achieve better performance in this application, an improved weakly supervised hierarchical deep learning pedestrian recognition algorithm with two-dimensional deep belief networks is proposed. The improvements are made by taking into consideration the weaknesses of structure and training methods of existing classifiers. First, traditional one-dimensional deep belief network is expanded to two-dimensional that allows image matrix to be loaded directly to preserve more information of a sample space. Then, a determination regularization term with small weight is added to the traditional unsupervised training objective function. By this modification, original unsupervised training is transformed to weakly supervised training. Subsequently, that gives the extracted features discrimination ability. Multiple sets of comparative experiments show that the performance of the proposed algorithm is better than other deep learning algorithms in recognition rate and outperforms most of the existing state-of-the-art methods in non-occlusion pedestrian data set while performs fair in weakly and heavily occlusion data set.

A method of locating the 3D centers of retroreflectors based on deep learning

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
BinBin Zhang ◽  
Fumin Zhang ◽  
Xinghua Qu
Keyword(s):  
Deep Learning ◽  
Data Augmentation ◽  
Measurement Techniques ◽  
Recognition Rate ◽  
Detection Algorithm ◽  
Large Field ◽  
Small Scale ◽  
Data Set ◽  
Content Type ◽  
Ellipse Detection

Purpose Laser-based measurement techniques offer various advantages over conventional measurement techniques, such as no-destructive, no-contact, fast and long measuring distance. In cooperative laser ranging systems, it’s crucial to extract center coordinates of retroreflectors to accomplish automatic measurement. To solve this problem, this paper aims to propose a novel method. Design/methodology/approach We propose a method using Mask RCNN (Region Convolutional Neural Network), with ResNet101 (Residual Network 101) and FPN (Feature Pyramid Network) as the backbone, to localize retroreflectors, realizing automatic recognition in different backgrounds. Compared with two other deep learning algorithms, experiments show that the recognition rate of Mask RCNN is better especially for small-scale targets. Based on this, an ellipse detection algorithm is introduced to obtain the ellipses of retroreflectors from recognized target areas. The center coordinates of retroreflectors in the camera coordinate system are obtained by using a mathematics method. Findings To verify the accuracy of this method, an experiment was carried out: the distance between two retroreflectors with a known distance of 1,000.109 mm was measured, with 2.596 mm root-mean-squar error, meeting the requirements of the coarse location of retroreflectors. Research limitations/implications The research limitations/implications are as follows: (i) As the data set only has 200 pictures, although we have used some data augmentation methods such as rotating, mirroring and cropping, there is still room for improvement in the generalization ability of detection. (ii) The ellipse detection algorithm needs to work in relatively dark conditions, as the retroreflector is made of stainless steel, which easily reflects light. Originality/value The originality/value of the article lies in being able to obtain center coordinates of multiple retroreflectors automatically even in a cluttered background; being able to recognize retroreflectors with different sizes, especially for small targets; meeting the recognition requirement of multiple targets in a large field of view and obtaining 3 D centers of targets by monocular model-based vision.

scholarly journals Real-Time Vehicle Make and Model Recognition with the Residual SqueezeNet Architecture

Sensors ◽  
2019 ◽  
Vol 19 (5) ◽  
pp. 982 ◽  
Author(s):  
Hyo Lee ◽  
Ihsan Ullah ◽  
Weiguo Wan ◽  
Yongbin Gao ◽  
Zhijun Fang
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Large Scale ◽  
Recognition Rate ◽  
Experimental Results ◽  
Learning Approach ◽  
Deep Model ◽  
Proposed Model ◽  
Real Time Applications ◽  
Model Recognition

Make and model recognition (MMR) of vehicles plays an important role in automatic vision-based systems. This paper proposes a novel deep learning approach for MMR using the SqueezeNet architecture. The frontal views of vehicle images are first extracted and fed into a deep network for training and testing. The SqueezeNet architecture with bypass connections between the Fire modules, a variant of the vanilla SqueezeNet, is employed for this study, which makes our MMR system more efficient. The experimental results on our collected large-scale vehicle datasets indicate that the proposed model achieves 96.3% recognition rate at the rank-1 level with an economical time slice of 108.8 ms. For inference tasks, the deployed deep model requires less than 5 MB of space and thus has a great viability in real-time applications.

scholarly journals Football Players’ Shooting Posture Norm Based on Deep Learning in Sports Event Video

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Guangliang Huang ◽  
Zhuangxu Lan ◽  
Guo Huang
Keyword(s):  
Deep Learning ◽  
Language Processing ◽  
Motion Tracking ◽  
Research Direction ◽  
Detection Algorithm ◽  
Recognition Algorithm ◽  
Target Motion ◽  
Football Players ◽  
Data Set ◽  
Offensive Tactics

Football is one of the favorite sports of people nowadays. Shooting is the ultimate goal of all offensive tactics in football matches. This is the most basic way to score a goal and the only way to score a goal. The choice and use of shooting technical indicators can have a great impact on the final result of the game. Therefore, how to improve the shooting technique of football players and how to adjust the shooting posture of football players are important issues faced by coaches and athletes. In recent years, deep learning has been widely used in various fields such as image classification and recognition and language processing. How to apply deep learning optimization to shooting gesture recognition is a very promising research direction. This article aims to study the football player’s shooting posture specification based on deep learning in sports event videos. Based on the analysis of target motion detection algorithm, target motion tracking algorithm, target motion recognition algorithm, and football shooting posture classification, KTH and Weizmann data sets are used. As the experimental verification data set of this article, the shooting posture of football players in the sports event video is recognized, and the accuracy of the action recognition is finally calculated to standardize the football shooting posture. The experimental results show that the Weizmann data set has a higher accuracy rate than the KTH data set and is more suitable for shooting attitude specifications.

scholarly journals Deep Learning based Arrhythmia Classification with an ECG Acquisition System

2019 ◽  
Vol 9 (2) ◽  
pp. 3849-3852
Keyword(s):  
Deep Learning ◽  
Premature Ventricular Contraction ◽  
Detection Algorithm ◽  
Raspberry Pi ◽  
Data Set ◽  
Bundle Branch Block ◽  
Ecg Signals ◽  
Peak Detection Algorithm ◽  
Different Types ◽  
Testing Accuracy

One of the issues that the human body faces is arrhythmia, a condition where the human heartbeat is either irregular, too slow or too fast. One of the ways to diagnose arrhythmia is by using ECG signals, the best diagnostic tool for detection of arrhythmia. This paper describes a deep learning approach to check whether signs of arrhythmia, in a given input signal, are present or not. A batch normalized CNN is used to classify the ECG signals based on the different types of arrhythmia. The model has achieved 96.39% training accuracy and 97% testing accuracy. The ECG signals are classified into five classes namely: Normal beats, Premature Ventricular Contraction (PVC) beats, Right Bundle Branch Block (RBBB) beats, Left Bundle Branch Block (LBBB) beats and Paced beats. A peak detection algorithm with six simple steps is designed to detect R-peaks from the ECG signals. A hardware device is built using Raspberry Pi to acquire ECG signals, which are then sent to the trained CNN for classification. The data-set for training is obtained from the MIT-BIH repository. Keras and Tensorflow libraries are used to design and develop the CNN and an application is designed using ’MEAN’ stack and ’Flask’ based servers.

scholarly journals A Target Detection Algorithm for Remote Sensing Images Based on Deep Learning

2021 ◽  
Vol 2021 ◽  
pp. 1-6
Author(s):  
Yi Lv ◽  
Zhengbo Yin ◽  
Zhezhou Yu
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
Target Detection ◽  
Detection Efficiency ◽  
False Positive Rate ◽  
Remote Sensing Image ◽  
Detection Algorithm ◽  
Detection Accuracy ◽  
Data Set ◽  
Detection Effect

In order to improve the accuracy of remote sensing image target detection, this paper proposes a remote sensing image target detection algorithm DFS based on deep learning. Firstly, dimension clustering module, loss function, and sliding window segmentation detection are designed. The data set used in the experiment comes from GoogleEarth, and there are 6 types of objects: airplanes, boats, warehouses, large ships, bridges, and ports. Training set, verification set, and test set contain 73490 images, 22722 images, and 2138 images, respectively. It is assumed that the number of detected positive samples and negative samples is A and B, respectively, and the number of undetected positive samples and negative samples is C and D, respectively. The experimental results show that the precision-recall curve of DFS for six types of targets shows that DFS has the best detection effect for bridges and the worst detection effect for boats. The main reason is that the size of the bridge is relatively large, and it is clearly distinguished from the background in the image, so the detection difficulty is low. However, the target of the boat is very small, and it is easy to be mixed with the background, so it is difficult to detect. The MAP of DFS is improved by 12.82%, the detection accuracy is improved by 13%, and the recall rate is slightly decreased by 1% compared with YOLOv2. According to the number of detection targets, the number of false positives (FPs) of DFS is much less than that of YOLOv2. The false positive rate is greatly reduced. In addition, the average IOU of DFS is 11.84% higher than that of YOLOv2. For small target detection efficiency and large remote sensing image detection, the DFS algorithm has obvious advantages.

scholarly journals Detection of Surface Defects in Logs Using Point Cloud Data and Deep Learning

2021 ◽  
Vol 15 ◽  
pp. 607-616
Author(s):  
Shengbo Liu ◽  
Pengyuan Fu ◽  
Lei Yan ◽  
Jian Wu ◽  
Yandong Zhao
Keyword(s):  
Deep Learning ◽  
Surface Defects ◽  
Recognition Rate ◽  
Point Clouds ◽  
Data Set ◽  
Cloud Data ◽  
Wood Processing ◽  
3D Point Clouds ◽  
3D Data ◽  
Deep Learning Network

Deep learning classification based on 3D point clouds has gained considerable research interest in recent years.The classification and quantitative analysis of wood defects are of great significance to the wood processing industry. In order to solve the problems of slow processing and low robustness of 3D data. This paper proposes an improvement based on littlepoint CNN lightweight deep learning network, adding BN layer. And based on the data set made by ourselves, the test is carried out. The new network bnlittlepoint CNN has been improved in speed and recognition rate. The correct rate of recognition for non defect log, non defect log and defect log as well as defect knot and dead knot can reach 95.6%.Finally, the "dead knot" and "loose knot" are quantitatively analyzed based on the "integral" idea, and the volume and surface area of the defect are obtained to a certain extent,the error is not more than 1.5% and the defect surface reconstruction is completed based on the triangulation idea.

scholarly journals Classification of Architectural Designs using Deep Learning

2020 ◽  
Vol 9 (3) ◽  
pp. 2471-2474
Keyword(s):  
Deep Learning ◽  
Recognition Rate ◽  
Automatic Identification ◽  
Architectural Style ◽  
Data Set ◽  
Mughal Empire ◽  
Scientific Methods ◽  
Architectural Styles ◽  
Bottle Neck

Architecture style of buildings play’s an important role in various aspects. Architectural style or the construction method affects the human health in multiple ways. Many dynasties are ruled India and constructed various types of monuments. So, In this proposed work popular dynasties like Hoysala dynasty, Vijayanagar empire, Mughal empire, Nizam’s of Hyderabad, Chalukya dynasty etc. are considered for creating dataset for the work. The architects of those times had really good knowledge about the different scientific methods to be used for construction. This project aims at classification of different architectural styles. Automatic identification of different architectural styles would facilitate different applications. The dataset is manually created by downloading images from various websites. Deep learning, inception v3 master algorithm are used. Experiments are performed using tenser flow and bottle neck files are created for validation. Good recognition rate is achieved with a fewer data set.

scholarly journals Network Abnormal Data Detection Based on Deep Learning Model

CONVERTER ◽  
2021 ◽  
pp. 64-73
Author(s):  
Yang Dong
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Detection System ◽  
Feature Learning ◽  
Original Data ◽  
Detection Algorithm ◽  
Training Data ◽  
Data Detection ◽  
Classification Problems ◽  
Data Set

To improve intrusion detection system performance,many algorithms are used to improve the performance of IDS systems, especially deep learning models. This paper presents an algorithm based on the model MLP, the training data set is the KDD99 data set, and the original data of the data set is vectorized by one-hot encoding, and the feature data is processed by Z-Score, and then the feature vector is encoded, and then the multi-layer perception is used The machine network performs feature learning, and finally trains the classifier model for detection. Traditional network anomaly detection algorithm models mainly use manual selection methods, and the accuracy and efficiency of classification problems are not high. This article first proposed the role of multilayer perceptron in Adam optimizer. The test of the KDD99 data set has been completed. The algorithm accuracy rate can reach 99%. For future network abnormal data detection work, an algorithm model that can realize real-time online detection is provided, which will have higher accuracy and better real-time performance.

scholarly journals A Demand-Side Load Event Detection Algorithm Based on Wide-Deep Neural Networks and Randomized Sparse Backpropagation

2021 ◽  
Vol 9 ◽  
Author(s):  
Chen Li ◽  
Gaoqi Liang ◽  
Huan Zhao ◽  
Guo Chen
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Deep Learning ◽  
Event Detection ◽  
Detection Algorithm ◽  
Demand Side ◽  
Rule Based ◽  
Detection Algorithms ◽  
Deep Model ◽  
Conventional Machine

Event detection is an important application in demand-side management. Precise event detection algorithms can improve the accuracy of non-intrusive load monitoring (NILM) and energy disaggregation models. Existing event detection algorithms can be divided into four categories: rule-based, statistics-based, conventional machine learning, and deep learning. The rule-based approach entails hand-crafted feature engineering and carefully calibrated thresholds; the accuracies of statistics-based and conventional machine learning methods are inferior to the deep learning algorithms due to their limited ability to extract complex features. Deep learning models require a long training time and are hard to interpret. This paper proposes a novel algorithm for load event detection in smart homes based on wide and deep learning that combines the convolutional neural network (CNN) and the soft-max regression (SMR). The deep model extracts the power time series patterns and the wide model utilizes the percentile information of the power time series. A randomized sparse backpropagation (RSB) algorithm for weight filters is proposed to improve the robustness of the standard wide-deep model. Compared to the standard wide-deep, pure CNN, and SMR models, the hybrid wide-deep model powered by RSB demonstrates its superiority in terms of accuracy, convergence speed, and robustness.

A Comparative Analysis of Deep Learning Approaches for Network Intrusion Detection Systems (N-IDSs)

2019 ◽  
Vol 11 (3) ◽  
pp. 65-89 ◽  
Author(s):  
Vinayakumar R ◽  
Soman KP ◽  
Prabaharan Poornachandran
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Intrusion Detection ◽  
Recurrent Neural Network ◽  
Network Architecture ◽  
Series Data ◽  
Network Intrusion Detection ◽  
Data Set ◽  
Deep Model ◽  
Network Intrusion

Recently, due to the advance and impressive results of deep learning techniques in the fields of image recognition, natural language processing and speech recognition for various long-standing artificial intelligence (AI) tasks, there has been a great interest in applying towards security tasks too. This article focuses on applying these deep taxonomy techniques to network intrusion detection system (N-IDS) with the aim to enhance the performance in classifying the network connections as either good or bad. To substantiate this to NIDS, this article models network traffic as a time series data, specifically transmission control protocol / internet protocol (TCP/IP) packets in a predefined time-window with a supervised deep learning methods such as recurrent neural network (RNN), identity matrix of initialized values typically termed as identity recurrent neural network (IRNN), long short-term memory (LSTM), clock-work RNN (CWRNN) and gated recurrent unit (GRU), utilizing connection records of KDDCup-99 challenge data set. The main interest is given to evaluate the performance of RNN over newly introduced method such as LSTM and IRNN to alleviate the vanishing and exploding gradient problem in memorizing the long-term dependencies. The efficient network architecture for all deep models is chosen based on comparing the performance of various network topologies and network parameters. The experiments of such chosen efficient configurations of deep models were run up to 1,000 epochs by varying learning-rates between 0.01-05. The observed results of IRNN are relatively close to the performance of LSTM on KDDCup-99 NIDS data set. In addition to KDDCup-99, the effectiveness of deep model architectures are evaluated on refined version of KDDCup-99: NSL-KDD and most recent one, UNSW-NB15 NIDS datasets.

Sign in / Sign up

Export Citation Format

Share Document