Existing almost deep learning methods rely on a large amount of annotated data, so they are inappropriate for forest fire smoke detection with limited data. In this paper, a novel hybrid attention-based few-shot learn...Existing almost deep learning methods rely on a large amount of annotated data, so they are inappropriate for forest fire smoke detection with limited data. In this paper, a novel hybrid attention-based few-shot learning method, named Attention-Based Prototypical Network, is proposed for forest fire smoke detection. Specifically, feature extraction network, which consists of convolutional block attention module, could extract high-level and discriminative features and further decrease the false alarm rate resulting from suspected smoke areas. Moreover, we design a metalearning module to alleviate the overfitting issue caused by limited smoke images, and the meta-learning network enables achieving effective detection via comparing the distance between the class prototype of support images and the features of query images. A series of experiments on forest fire smoke datasets and miniImageNet dataset testify that the proposed method is superior to state-of-the-art few-shot learning approaches.展开更多
In this study,PCR-denaturing gradient gel electrophoresis(DGGE)was applied to analyze the microbial communities in lake sediments from Lake Xuanwu,Lake Mochou in Nanjing and Lake Taihu in Wuxi.Sediment samples from se...In this study,PCR-denaturing gradient gel electrophoresis(DGGE)was applied to analyze the microbial communities in lake sediments from Lake Xuanwu,Lake Mochou in Nanjing and Lake Taihu in Wuxi.Sediment samples from seven locations in three lakes were collected and their genomic DNAs were extracted.The DNA yields of the sediments of Lake Xuanwu and Lake Mochou were high(10 mg/g),while that of sediments in Lake Taihu was relatively low.After DNA purification,the 16S rDNA genes(V3 to V5 region)were amplified and the amplified DNA fragments were separated by parallel DGGE.The DGGE profiles showed that there were five common bands in all the lake sediment samples indicating that there were similarities among the populations of microorganisms in all the lake sediments.The DGGE profiles of Lake Xuanwu and Lake Mochou were similar and about 20 types of micro-organisms were identified in the sediment samples of both lakes.These results suggest that the sediment samples of these two city lakes(Xuanwu,Mochou)have similar microbial communities.However,the DGGE profiles of sediment samples in Lake Taihu were significantly different from these two lakes.Furthermore,the DGGE profiles of sediment samples in different locations in Lake Taihu were also different,suggesting that the microbial communities in Lake Taihu are more diversified than those in Lake Xuanwu and Lake Mochou.The differences in microbial diversity may be caused by the different environmental conditions,such as redox potential,pH,and the concentrations of organic matters.Seven major bands of 16S rDNA genes fragments from the DGGE profiles of sediment samples were further reamplified and sequenced.The results of sequencing analysis indicate that five sequences shared 99%-100%homology with known sequences(Bacillus and Brevibacillus,uncultured bacteria),while the other two sequences shared 93%-96%homology with known sequences(Acinetobacter,and Bacillus).The study shows that the PCR-DGGE technique combined with sequence analysis is a feasible and efficient method for the determination of microbial communities in sediment samples.展开更多
基金The work was supported by the National Key R&D Program of China(Grant No.2020YFC1511601)Fundamental Research Funds for the Central Universities(Grant No.2019SHFWLC01).
文摘Existing almost deep learning methods rely on a large amount of annotated data, so they are inappropriate for forest fire smoke detection with limited data. In this paper, a novel hybrid attention-based few-shot learning method, named Attention-Based Prototypical Network, is proposed for forest fire smoke detection. Specifically, feature extraction network, which consists of convolutional block attention module, could extract high-level and discriminative features and further decrease the false alarm rate resulting from suspected smoke areas. Moreover, we design a metalearning module to alleviate the overfitting issue caused by limited smoke images, and the meta-learning network enables achieving effective detection via comparing the distance between the class prototype of support images and the features of query images. A series of experiments on forest fire smoke datasets and miniImageNet dataset testify that the proposed method is superior to state-of-the-art few-shot learning approaches.
基金This study is supported by the National Basic Research Program(2002CB412307)the National High Technology Research and Development Program of China(Water Environment Program)(2002AA601011).
文摘In this study,PCR-denaturing gradient gel electrophoresis(DGGE)was applied to analyze the microbial communities in lake sediments from Lake Xuanwu,Lake Mochou in Nanjing and Lake Taihu in Wuxi.Sediment samples from seven locations in three lakes were collected and their genomic DNAs were extracted.The DNA yields of the sediments of Lake Xuanwu and Lake Mochou were high(10 mg/g),while that of sediments in Lake Taihu was relatively low.After DNA purification,the 16S rDNA genes(V3 to V5 region)were amplified and the amplified DNA fragments were separated by parallel DGGE.The DGGE profiles showed that there were five common bands in all the lake sediment samples indicating that there were similarities among the populations of microorganisms in all the lake sediments.The DGGE profiles of Lake Xuanwu and Lake Mochou were similar and about 20 types of micro-organisms were identified in the sediment samples of both lakes.These results suggest that the sediment samples of these two city lakes(Xuanwu,Mochou)have similar microbial communities.However,the DGGE profiles of sediment samples in Lake Taihu were significantly different from these two lakes.Furthermore,the DGGE profiles of sediment samples in different locations in Lake Taihu were also different,suggesting that the microbial communities in Lake Taihu are more diversified than those in Lake Xuanwu and Lake Mochou.The differences in microbial diversity may be caused by the different environmental conditions,such as redox potential,pH,and the concentrations of organic matters.Seven major bands of 16S rDNA genes fragments from the DGGE profiles of sediment samples were further reamplified and sequenced.The results of sequencing analysis indicate that five sequences shared 99%-100%homology with known sequences(Bacillus and Brevibacillus,uncultured bacteria),while the other two sequences shared 93%-96%homology with known sequences(Acinetobacter,and Bacillus).The study shows that the PCR-DGGE technique combined with sequence analysis is a feasible and efficient method for the determination of microbial communities in sediment samples.
