This page has only limited features, please log in for full access.
Desert segmentation of remote sensing images is the basis of analysis of desert area. Desert images are usually characterized by large image size, large-scale change, and irregular location distribution of surface objects. The multi-scale fusion method is widely used in the existing deep learning segmentation models to solve the above problems. Based on the idea of multi-scale feature extraction, this paper took the segmentation results of each scale as an independent optimization task and proposed a multi-resolution supervision network (MrsSeg) to further improve the desert segmentation result. Due to the different optimization difficulty of each branch task, we also proposed an auxiliary adaptive weighted loss function (AWL) to automatically optimize the training process. MrsSeg first used a lightweight backbone to extract different-resolution features, then adopted a multi-resolution fusion module to fuse the local information and global information, and finally, a multi-level fusion decoder was used to aggregate and merge the features at different levels to get the desert segmentation result. In this method, each branch loss was treated as an independent task, AWL was proposed to calculate and adjust the weight of each branch. By giving priority to the easy tasks, the improved loss function could effectively improve the convergence speed of the model and the desert segmentation result. The experimental results showed that MrsSeg-AWL effectively improved the learning ability of the model and has faster convergence speed, lower parameter complexity, and more accurate segmentation results.
Lexuan Wang; Liguo Weng; Min Xia; Jia Liu; Haifeng Lin. Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation. Remote Sensing 2021, 13, 2054 .
AMA StyleLexuan Wang, Liguo Weng, Min Xia, Jia Liu, Haifeng Lin. Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation. Remote Sensing. 2021; 13 (11):2054.
Chicago/Turabian StyleLexuan Wang; Liguo Weng; Min Xia; Jia Liu; Haifeng Lin. 2021. "Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation." Remote Sensing 13, no. 11: 2054.
Building and road extraction from remote sensing images is of great significance to urban planning. At present, most of building and road extraction models adopt deep learning semantic segmentation method. However, the existing semantic segmentation methods did not pay enough attention to the feature information between hidden layers, which led to the neglect of the category of context pixels in pixel classification, resulting in these two problems of large-scale misjudgment of buildings and disconnection of road extraction. In order to solve these problem, this paper proposes a Non-Local Feature Search Network (NFSNet) that can improve the segmentation accuracy of remote sensing images of buildings and roads, and to help achieve accurate urban planning. By strengthening the exploration of hidden layer feature information, it can effectively reduce the large area misclassification of buildings and road disconnection in the process of segmentation. Firstly, a Self-Attention Feature Transfer (SAFT) module is proposed, which searches the importance of hidden layer on channel dimension, it can obtain the correlation between channels. Secondly, the Global Feature Refinement (GFR) module is introduced to integrate the features extracted from the backbone network and SAFT module, it enhances the semantic information of the feature map and obtains more detailed segmentation output. The comparative experiments demonstrate that the proposed method outperforms state-of-the-art methods, and the model complexity is the lowest.
Cheng Ding; Liguo Weng; Min Xia; Haifeng Lin. Non-Local Feature Search Network for Building and Road Segmentation of Remote Sensing Image. ISPRS International Journal of Geo-Information 2021, 10, 245 .
AMA StyleCheng Ding, Liguo Weng, Min Xia, Haifeng Lin. Non-Local Feature Search Network for Building and Road Segmentation of Remote Sensing Image. ISPRS International Journal of Geo-Information. 2021; 10 (4):245.
Chicago/Turabian StyleCheng Ding; Liguo Weng; Min Xia; Haifeng Lin. 2021. "Non-Local Feature Search Network for Building and Road Segmentation of Remote Sensing Image." ISPRS International Journal of Geo-Information 10, no. 4: 245.
Due to the various shapes, textures, and colors of fires, forest fire detection is a challenging task. The traditional image processing method relies heavily on manmade features, which is not universally applicable to all forest scenarios. In order to solve this problem, the deep learning technology is applied to learn and extract features of forest fires adaptively. However, the limited learning and perception ability of individual learners is not sufficient to make them perform well in complex tasks. Furthermore, learners tend to focus too much on local information, namely ground truth, but ignore global information, which may lead to false positives. In this paper, a novel ensemble learning method is proposed to detect forest fires in different scenarios. Firstly, two individual learners Yolov5 and EfficientDet are integrated to accomplish fire detection process. Secondly, another individual learner EfficientNet is responsible for learning global information to avoid false positives. Finally, detection results are made based on the decisions of three learners. Experiments on our dataset show that the proposed method improves detection performance by 2.5% to 10.9%, and decreases false positives by 51.3%, without any extra latency.
Renjie Xu; Haifeng Lin; Kangjie Lu; Lin Cao; Yunfei Liu. A Forest Fire Detection System Based on Ensemble Learning. Forests 2021, 12, 217 .
AMA StyleRenjie Xu, Haifeng Lin, Kangjie Lu, Lin Cao, Yunfei Liu. A Forest Fire Detection System Based on Ensemble Learning. Forests. 2021; 12 (2):217.
Chicago/Turabian StyleRenjie Xu; Haifeng Lin; Kangjie Lu; Lin Cao; Yunfei Liu. 2021. "A Forest Fire Detection System Based on Ensemble Learning." Forests 12, no. 2: 217.
Wildfire is a sudden and hazardous natural disaster. Currently, many schemes based on optical spectrum analysis have been proposed to detect wildfire, but obstacles in forest areas can decrease the efficiency of spectral monitoring, resulting in a wildfire detection system not being able to monitor the occurrence of wildfire promptly. In this paper, we propose a novel wildfire detection system using sound spectrum analysis based on the Internet of Things (IoT), which utilizes a wireless acoustic detection system to probe wildfire and distinguish the difference in the sound between the crown and the surface fire. We also designed a new power supply unit: tree-energy device, which utilizes the biological energy of the living trees to generate electricity. We implemented sound spectrum analysis on the data collected by sound sensors and then combined our classification algorithms. The results describe that the sound frequency of the crown fire is about 0–400 Hz, while the sound frequency of the surface fire ranges from 0 to 15,000 Hz. However, the accuracy of the classification method is affected by some factors, such as the distribution of sensors, the loss of energy in sound transmission, and the delay of data transmission. In the simulation experiments, the recognition rate of the method can reach about 70%.
Shuo Zhang; Demin Gao; Haifeng Lin; Quan Sun. Wildfire Detection Using Sound Spectrum Analysis Based on the Internet of Things. Sensors 2019, 19, 5093 .
AMA StyleShuo Zhang, Demin Gao, Haifeng Lin, Quan Sun. Wildfire Detection Using Sound Spectrum Analysis Based on the Internet of Things. Sensors. 2019; 19 (23):5093.
Chicago/Turabian StyleShuo Zhang; Demin Gao; Haifeng Lin; Quan Sun. 2019. "Wildfire Detection Using Sound Spectrum Analysis Based on the Internet of Things." Sensors 19, no. 23: 5093.
The species of genus Actinidia are economically and nutritionally important fruits with remarkably high vitamin C content. Here, we assembled and characterized the complete chloroplast (cp) genome sequence of Actinidia fulvicoma (A. fulvicoma) using Illumina paired-end sequencing data. The cp genome is 157,339 bp in length, including a large single-copy region (LSC) of 88,741 bp, a small single-copy region (SSC) of 20,512 bp, and a pair of 24,043 bp inverted repeat (IR) regions. A total of 131 genes, consisting of 85 protein-coding genes, 38 tRNA genes, and 8 rRNA genes, were annotated in the A. fulvicoma cp genome. Phylogenetic analysis confirmed the evolutionary position of A. fulvicoma within the genus Actinidia.
Fuquan Zhang; Zhihao Yan; Yiqing Xu; Haifeng Lin. The complete chloroplast genome of Actinidia fulvicoma. Mitochondrial DNA Part B 2019, 4, 4089 -4090.
AMA StyleFuquan Zhang, Zhihao Yan, Yiqing Xu, Haifeng Lin. The complete chloroplast genome of Actinidia fulvicoma. Mitochondrial DNA Part B. 2019; 4 (2):4089-4090.
Chicago/Turabian StyleFuquan Zhang; Zhihao Yan; Yiqing Xu; Haifeng Lin. 2019. "The complete chloroplast genome of Actinidia fulvicoma." Mitochondrial DNA Part B 4, no. 2: 4089-4090.
In Rechargeable Wireless Sensor Networks(R-WSNs), it is critical for data collection because a sensor has to operate in a very low and dynamic duty cycle owing to sporadic availability of energy. In this work, we propose a distribute maximum rate allocation based on data aggregation to compute an upper data generation rate by maximizing it as a linear programming problem. Subsequently, a dual problem by introducing Lagrange multipliers is constructed, and subgradient algorithms are used to solve it in a distributed manner. The resulting algorithms are guaranteed to converge to an optimal value with low computational complexity. Through extensive simulation and experiments, we demonstrate our algorithm is efficient to maximize data collection rate in rechargeable wireless sensor networks.
Demin Gao; Jinchi Zhang; Fuquan Zhang; Haifeng Lin. Distributed Optimal Maximum Rate Allocation based on Data Aggregation in Rechargeable Wireless Sensor Networks. International Journal of Online Engineering (iJOE) 2018, 14, 172 -179.
AMA StyleDemin Gao, Jinchi Zhang, Fuquan Zhang, Haifeng Lin. Distributed Optimal Maximum Rate Allocation based on Data Aggregation in Rechargeable Wireless Sensor Networks. International Journal of Online Engineering (iJOE). 2018; 14 (3):172-179.
Chicago/Turabian StyleDemin Gao; Jinchi Zhang; Fuquan Zhang; Haifeng Lin. 2018. "Distributed Optimal Maximum Rate Allocation based on Data Aggregation in Rechargeable Wireless Sensor Networks." International Journal of Online Engineering (iJOE) 14, no. 3: 172-179.
In Rechargeable Wireless Sensor Networks (R-WSNs), in order to achieve the maximum data collection rate it is critical that sensors operate in very low duty cycles because of the sporadic availability of energy. A sensor has to stay in a dormant state in most of the time in order to recharge the battery and use the energy prudently. In addition, a sensor cannot always conserve energy if a network is able to harvest excessive energy from the environment due to its limited storage capacity. Therefore, energy exploitation and energy saving have to be traded off depending on distinct application scenarios. Since higher data collection rate or maximum data collection rate is the ultimate objective for sensor deployment, surplus energy of a node can be utilized for strengthening packet delivery efficiency and improving the data generating rate in R-WSNs. In this work, we propose an algorithm based on data aggregation to compute an upper data generation rate by maximizing it as an optimization problem for a network, which is formulated as a linear programming problem. Subsequently, a dual problem by introducing Lagrange multipliers is constructed, and subgradient algorithms are used to solve it in a distributed manner. At the same time, a topology controlling scheme is adopted for improving the network’s performance. Through extensive simulation and experiments, we demonstrate that our algorithm is efficient at maximizing the data collection rate in rechargeable wireless sensor networks.
Haifeng Lin; Di Bai; Demin Gao; Yunfei Liu. Maximum Data Collection Rate Routing Protocol Based on Topology Control for Rechargeable Wireless Sensor Networks. Sensors 2016, 16, 1201 .
AMA StyleHaifeng Lin, Di Bai, Demin Gao, Yunfei Liu. Maximum Data Collection Rate Routing Protocol Based on Topology Control for Rechargeable Wireless Sensor Networks. Sensors. 2016; 16 (8):1201.
Chicago/Turabian StyleHaifeng Lin; Di Bai; Demin Gao; Yunfei Liu. 2016. "Maximum Data Collection Rate Routing Protocol Based on Topology Control for Rechargeable Wireless Sensor Networks." Sensors 16, no. 8: 1201.
In this paper, a routing protocol for k-anycast communication based upon the anycast tree scheme is proposed for wireless sensor networks. Multiple-metrics are utilized for instructing the route discovery. A source initiates to create a spanning tree reaching any one sink with source node as the root. Subsequently, we introduce three schemes for k-anycast: a packet is transmitted to exact k sinks, at least k sinks, and at most k sinks, where, a packet can be transmitted to k or more than k sinks benefiting from broadcast technique without wasting the source energy for replicating it. k-anycast communication technique is utilized for enhancing transmission reliability, load-balancing and security purpose by collecting multiple copies of a packet from a source and verifying the information of monitoring field. In r-WSNs, due to energy replenished continually and limited energy storage capacity, a sensor cannot be always beneficial to conserve energy when a network can harvest excessive energy from the environment. Therefore, the surplus energy of sensor can be used for strengthening data transmission. In this paper, a routing protocol for k-anycast communication based upon the anycast tree scheme is proposed for wireless sensor networks.Multiple-metrics are utilized for instructing the route discovery. A source initiates to create a spanning tree reaching any one sink with source node as the root.We introduce three schemes for k-anycast: a packet is transmitted to exact k sinks, at least k sinks, and at most k sinks, where, a packet can be transmitted to k or more than k sinks benefiting from broadcast technique without wasting the source energy for replicating it. Through extensive simulation and experiments, we demonstrate our algorithm is feasible and efficient to collect multiple copies of a packet from a source in rechargeable wireless sensor networks.
Demin Gao; Haifeng Lin; Xiaofeng Liu. Routing protocol for k-anycast communication in rechargeable wireless sensor networks. Computer Standards & Interfaces 2016, 43, 12 -20.
AMA StyleDemin Gao, Haifeng Lin, Xiaofeng Liu. Routing protocol for k-anycast communication in rechargeable wireless sensor networks. Computer Standards & Interfaces. 2016; 43 ():12-20.
Chicago/Turabian StyleDemin Gao; Haifeng Lin; Xiaofeng Liu. 2016. "Routing protocol for k-anycast communication in rechargeable wireless sensor networks." Computer Standards & Interfaces 43, no. : 12-20.