Pedestrian and cyclist detection based on deep neural network fast R-CNN

In this article, a unified joint detection framework for pedestrian and cyclist is established to realize the joint detection of pedestrian and cyclist targets. Based on the target detection of fast regional convolution neural network, a deep neural network model suitable for pedestrian and cyclist detection is established. Experiments for poor detection results for small-sized targets and complex and changeable background environment; various network improvement schemes such as difficult case extraction, multilayer feature fusion, and multitarget candidate region input were designed to improve detection and to solve the problems of frequent false detections and missed detections in pedestrian and cyclist target detection. Results of experimental verification of the pedestrian and cyclist database established in Beijing’s urban traffic environment showed that the proposed joint detection method for pedestrians and cyclists can realize the stable tracking of joint detection and clearly distinguish different target categories. Therefore, an important basis for the behavior decision of intelligent vehicles is provided.

Download Full-text

Multi-local Feature Target Detection Method Based on Deep Neural Network

Advances in Intelligent Systems and Computing - Genetic and Evolutionary Computing ◽

10.1007/978-981-13-5841-8_52 ◽

2019 ◽

pp. 501-510

Author(s):

Guojie Li ◽

Wenxue Wei ◽

Wen Sun

Keyword(s):

Neural Network ◽

Target Detection ◽

Deep Neural Network ◽

Detection Method ◽

Local Feature ◽

Feature Target

Download Full-text

Research on the application of target detection based on deep learning technology in power grid operation inspection

E3S Web of Conferences ◽

10.1051/e3sconf/202123601035 ◽

2021 ◽

Vol 236 ◽

pp. 01035

Author(s):

Peng Weifu ◽

Du Shu ◽

Chen Shaolei ◽

Zhou Qing ◽

Tang Na

Keyword(s):

Neural Network ◽

Deep Learning ◽

Power System ◽

Target Detection ◽

Early Warning ◽

Deep Neural Network ◽

Detection Method ◽

Video Data ◽

Detection Speed ◽

External Damage

-External damage to power facilities caused by crane, excavator and other construction operations increases year by year, which will seriously threaten the safe operation of power system. It is an important measure to ensure the safe and reliable operation of power system to implement intelligent monitoring and early warning of power external breakdown through video and other non-contact observation means. The video data of power mainly comes from the fixed monitoring of helicopters, uavs and transformation poles and towers, which is characterized by large amount of data, complex scenes and serious environmental interference. The traditional target detection method usually selects the candidate area first, and then makes judgment based on the characteristics of human construction. The detection speed is slow and the accuracy is low, which makes it impossible to monitor the video data in real time, so as to make timely and accurate early warning and intervention fbr external damage. The target detection method based on deep learning optimizes or even eliminates the selection of candidate regions, which greatly speeds up the detection speed. By learning a lot of target samples through the deep neural network, the characteristics of high robustness are gradually fitted to make the target judgment more accurate. There are three key problems in introducing the target detection method based on deep learning into the power video detection: Firstly, the target detection method based on deep learning has a large amount of calculation and many parameters. In order to realize in-place operation on terminals with limited computing and storage capacity, it is necessary to find a practical method to simplify the network and reduce the amount of operational data in the detection process, which is the key to realize in-place operation and terminal operation of deep neural network. Secondly, for specific application scenarios, the effect of different target detection algorithms varies greatly, and there is a strong particularity of power video. Finding an effective target detection method is the key to improve the detection speed and accuracy. Finally, with the continuous development of deep learning, the structure of deep neural network changes with each passing day, and each has its own characteristics, which network structure is used as the feature extraction layer of target detection algorithm is the focus of research.

Download Full-text

A Method of Fast Segmentation for Banana Stalk Exploited Lightweight Multi-Feature Fusion Deep Neural Network

Machines ◽

10.3390/machines9030066 ◽

2021 ◽

Vol 9 (3) ◽

pp. 66

Author(s):

Tianci Chen ◽

Rihong Zhang ◽

Lixue Zhu ◽

Shiang Zhang ◽

Xiaomin Li

Keyword(s):

Neural Network ◽

Execution Time ◽

Deep Neural Network ◽

Feature Fusion ◽

Frame Rate ◽

Light Conditions ◽

Segmentation Method ◽

Segmentation Accuracy ◽

Automatic Picking ◽

Changing Light

In an orchard environment with a complex background and changing light conditions, the banana stalk, fruit, branches, and leaves are very similar in color. The fast and accurate detection and segmentation of a banana stalk are crucial to realize the automatic picking using a banana picking robot. In this paper, a banana stalk segmentation method based on a lightweight multi-feature fusion deep neural network (MFN) is proposed. The proposed network is mainly composed of encoding and decoding networks, in which the sandglass bottleneck design is adopted to alleviate the information a loss in high dimension. In the decoding network, a different sized dilated convolution kernel is used for convolution operation to make the extracted banana stalk features denser. The proposed network is verified by experiments. In the experiments, the detection precision, segmentation accuracy, number of parameters, operation efficiency, and average execution time are used as evaluation metrics, and the proposed network is compared with Resnet_Segnet, Mobilenet_Segnet, and a few other networks. The experimental results show that compared to other networks, the number of network parameters of the proposed network is significantly reduced, the running frame rate is improved, and the average execution time is shortened.

Download Full-text

Spectral Convolution Feature-Based SPD Matrix Representation for Signal Detection Using a Deep Neural Network

Entropy ◽

10.3390/e22090949 ◽

2020 ◽

Vol 22 (9) ◽

pp. 949

Author(s):

Jiangyi Wang ◽

Min Liu ◽

Xinwu Zeng ◽

Xiaoqiang Hua

Keyword(s):

Neural Network ◽

Signal Detection ◽

Convolutional Neural Network ◽

Deep Neural Network ◽

Detection Method ◽

Learning Algorithm ◽

Simulated Data ◽

Data Sets ◽

Feature Maps ◽

Simulated Data Sets

Convolutional neural networks have powerful performances in many visual tasks because of their hierarchical structures and powerful feature extraction capabilities. SPD (symmetric positive definition) matrix is paid attention to in visual classification, because it has excellent ability to learn proper statistical representation and distinguish samples with different information. In this paper, a deep neural network signal detection method based on spectral convolution features is proposed. In this method, local features extracted from convolutional neural network are used to construct the SPD matrix, and a deep learning algorithm for the SPD matrix is used to detect target signals. Feature maps extracted by two kinds of convolutional neural network models are applied in this study. Based on this method, signal detection has become a binary classification problem of signals in samples. In order to prove the availability and superiority of this method, simulated and semi-physical simulated data sets are used. The results show that, under low SCR (signal-to-clutter ratio), compared with the spectral signal detection method based on the deep neural network, this method can obtain a gain of 0.5–2 dB on simulated data sets and semi-physical simulated data sets.

Download Full-text