Forest fire identification based on Sparse-DenseNet model

Zhou Lang; Fan Kun; Qu Hua; Lü Yuanyuan; Zhang Zhengyi

doi:10.12171/j.1000-1522.20190371

Journal of Beijing Forestry University > 2020 > 42(10): 36-44. > DOI: 10.12171/j.1000-1522.20190371

Zhou Lang, Fan Kun, Qu Hua, Lü Yuanyuan, Zhang Zhengyi. Forest fire identification based on Sparse-DenseNet model[J]. Journal of Beijing Forestry University, 2020, 42(10): 36-44. DOI: 10.12171/j.1000-1522.20190371

Citation:

PDF (1108 KB)

Forest fire identification based on Sparse-DenseNet model

School of Economics and Management, Beijing Forestry University, Beijing 100083, China

More Information

Received Date: September 24, 2019
Revised Date: November 13, 2019
Available Online: September 08, 2020
Published Date: October 24, 2020

Graphical Abstract

Abstract

Abstract

Objective The frequent occurrence of forest fire brings great difficulty to forest prevention and control. Traditional forest fire identification algorithms have problems such as low accuracy and insufficient processing efficiency. At the same time, forest fire image data itself has strong complexity. It is necessary to comprehensively consider the recognition accuracy and generalization ability. For this reason, this paper uses the Sparse-DenseNet model to carry out the identification research of forest fire.
Method Firstly, the DenseNet model was sparsely transformed, and the sparse effect was generated by randomly shielding the nodes in the Dense Block module, so that the algorithm had the advantages of reducing over-fitting, mitigating gradient disappearance and accelerating convergence speed. Secondly, when image acquisition was carried out in the forest area, due to the relative motion between the camera device and the object being collected and the effect of light and shadow, the image data may be disturbed. Therefore, this paper uses python image processing tools to transform the image. The image dataset was expanded accordingly to fit the actual application scenario. Finally, this paper compares the performance of the Sparse-DenseNet model with other classical deep learning models on the forest fire dataset and the cifar10 dataset, and observes its effects.
Result The Sparse-DenseNet model had the characteristics of being lighter in structure, training faster, better effect of avoiding over-fitting. It had a good performance in the forest fire data set and the standard data set cifar10.
Conclusion The Sparse-DenseNet model proposed in this paper can effectively optimize the problems existing in traditional forest models and achieve good recognition results. The accuracy rate can reach 99.33%, which is better than DesenNet’s 98.15%, and the training time is only about three-quarters of DenseNet in same round.
- deep learning,
- DenseNet,
- sparsity,
- forest fire,
- image identification

FullText(HTML)

References (24)

References

[1]	马岩, 张明松, 杨春梅, 等. 森林火灾的危害及重要灭火手段的分析[J]. 森林工程, 2013, 29(6):25−27. Ma Y, Zhang M S, Yang C M, et al. Analysis of forest fire hazards and important means of fighting fires[J]. Forest Engineering, 2013, 29(6): 25−27.
[2]	林燕, 岩糯香. 森林防火现状及火灾控制措施[J]. 现代园艺, 2017(2):223. Lin Y, Yan N X. Current situation of forest fire prevention and fire control measures[J]. Xiandai Horticulture, 2017(2): 223.
[3]	邹全程, 王丽娜. 我国县(区)级单位森林防火能力建设研究: 以威海市环翠区为例[J]. 林业调查规划, 2017, 42(6):33−37. Zou Q C, Wang L N. Study on the capacity construction of forest fire prevention of counties (districts) in China: a case study of Huancui District, Weihai City[J]. Forest Inventory and Planning, 2017, 42(6): 33−37.
[4]	孙立研, 刘美玲, 周礼祥, 等. 基于气象因子深度学习的森林火灾预测方法[J]. 林业工程学报, 2019, 4(3):132−136. Sun L Y, Liu M L, Zhou L X, et al. Research on forest fire prediction method based on deep learning[J]. Journal of Forestry Engineering, 2019, 4(3): 132−136.
[5]	张军阳, 王慧丽, 郭阳, 等. 深度学习相关研究综述[J]. 计算机应用研究, 2018, 35(7):1921−1928, 1936. Zhang J Y, Wang H L, Guo Y, et al. Review of deep learning[J]. Application Research of Computers, 2018, 35(7): 1921−1928, 1936.
[6]	LeCun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521: 436−444. doi: 10.1038/nature14539.
[7]	李卫. 深度学习在图像识别中的研究及应用[D]. 武汉: 武汉理工大学, 2014. Li W. Research and application of deep learning in image recognition[D]. Wuhan: Wuhan University of Technology, 2014.
[8]	梁万杰, 曹宏鑫. 基于卷积神经网络的水稻虫害识别[J]. 江苏农业科学, 2017, 45(20):241−243, 253. Liang W J, Cao H X. Rice pest identification based on convolutional neural network[J]. Jiangsu Agricultural Sciences, 2017, 45(20): 241−243, 253.
[9]	Mohanty S P, Hughes D P, Salathé M. Using deep learning for image-based plant disease detection[J]. Frontiers in Plant Science, 2016, 7: 1−10. doi: 10.3389/fpls.2016.01419.
[10]	刘恒旭, 刘凯. 森林防火信息化建设技术问题与措施[J]. 科学技术创新, 2017(1):280. Liu H X, Liu K. Technical problems and measures for forest fire prevention information construction[J]. Science and Technology Innovation, 2017(1): 280.
[11]	傅天驹, 郑嫦娥, 田野, 等. 复杂背景下基于深度卷积神经网络的森林火灾识别[J]. 计算机与现代化, 2016(3):52−57. Fu T J, Zheng C E, Tian Y, et al. Forest fire recognition based on deep convolutional neural network under complex background[J]. Computer and Modernization, 2016(3): 52−57.
[12]	赵亚琴. 基于模糊神经网络的火灾识别算法[J]. 计算机仿真, 2015, 32(2):369−373. Zhao Y Q. Forest fire recognition algorithm based on fuzzy nerual network[J]. Computer Simulation, 2015, 32(2): 369−373.
[13]	李诚, 唐李洋, 潘李伟. 城镇森林交界域视频烟雾检测算法[J]. 计算机工程, 2018, 44(1):258−262. Li C, Tang L Y, Pan L W. Video smoke detection algorithm for wildland-urban interface[J]. Computer Engineering, 2018, 44(1): 258−262.
[14]	卢柯楠, 胡晓惠, 李小涛. 安防大数据技术与应用:基于多光谱大数据分析的智能森林火警检测[J]. 中国安全防范认证, 2017(3):10−15. Lu K N, Hu X H, Li X T. Security big data technology and application: intelligent forest fire alarm detection based on multispectral big data analysis[J]. China Security Protection Certification, 2017(3): 10−15.
[15]	Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C/OL]//Proceedings of the 32nd International Conference on Machine Learning. 2015: 448−456 [2019−05−06]. https://arxiv.org/pdf/1502.03167.pdf.
[16]	Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 2261−2269.
[17]	Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313: 504−507. doi: 10.1126/science.1127647.
[18]	Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 2818−2826.
[19]	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision & Pattern Recognition. Las Vegas: IEEE, 2016: 770−778.
[20]	Liu B Y, Wang M, Foroosh H, et al. Sparse convolutional neural networks[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015: 806−814.
[21]	Xu Q, Pan G. SparseConnect: regularising CNNs on fully connected layers[J]. Electronics Letters, 2017, 53(18): 1246−1248. doi: 10.1049/el.2017.2621
[22]	Rodriguez-Galiano V F, Ghimire B, Rogan J, et al. An assessment of the effectiveness of a random forest classifier for land-cover classification[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012, 67: 93−104. doi: 10.1016/j.isprsjprs.2011.11.002.
[23]	Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]//Advances in neural information processing systems 25. Lake Tahoe: Curran Associates, Inc., 2012: 1106−1114.
[24]	王红霞, 周家奇, 辜承昊, 等. 用于图像分类的卷积神经网络中激活函数的设计[J]. 浙江大学学报(工学版), 2019, 53(7):1363−1373. Wang H X, Zhou J Q, Gu C H, et al. Design of activation function in CNN for image classification[J]. Journal of Zhejiang University (Engineering Science), 2019, 53(7): 1363−1373.

[1]	Deng Wenwen, Dai Ying, Zhao Xiuhai, Zhang Chunyu. Biomass stability and influencing factors of mixed coniferous and broadleaved forests at different scales in Jiaohe, Jilin Province of northeastern China[J]. Journal of Beijing Forestry University, 2024, 46(7): 55-66. DOI: 10.12171/j.1000-1522.20220315
[2]	Yang Dongye, Yu Xinxiao, Li Xuhong, Jiang Tao, Jia Guodong. Characteristics of sap flow of degraded Populus simonii in Bashang Area, Hebei Province of northern China and its response to environmental factors[J]. Journal of Beijing Forestry University, 2024, 46(7): 36-43. DOI: 10.12171/j.1000-1522.20230332
[3]	Zheng Dongsheng, Liu Qijing. Effects of environmental factors on forest community distribution in Changbai Mountain Nature Reserve of northeastern China[J]. Journal of Beijing Forestry University, 2023, 45(8): 57-64. DOI: 10.12171/j.1000-1522.20220086
[4]	Ma Jing, Guo Jianbin, Liu Zebin, Wang Yanhui, Zhang Ziyou. Diurnal variations of stand transpiration of Larix principis-rupprechtii forest and its response to environmental factors in Liupan Mountains of northwestern China[J]. Journal of Beijing Forestry University, 2020, 42(5): 1-11. DOI: 10.12171/j.1000-1522.20190468
[5]	Luo Guisheng, Ma Lüyi, Jia Zhongkui, Wu Danni, Chi Mingfeng, Zhang Shumin, Zhao Guijuan. Correlation analysis between natural regeneration and environment in canopy gap of Chinese pine (Pinus tabuliformis) plantation[J]. Journal of Beijing Forestry University, 2019, 41(9): 59-68. DOI: 10.13332/j.1000-1522.20180416
[6]	Liu Wei, Wei Tianxing, Zhu Qingke. Dynamics of sap flow density of Populus hopeiensis and its responses to environmental variables in the water-wind erosion crisscross region on the Loess Plateau[J]. Journal of Beijing Forestry University, 2018, 40(5): 73-81. DOI: 10.13332/j.1000-1522.20180003
[7]	LI Xiao-mei, ZHANG Qiu-liang. Impact of climate factors on CO2 flux characteristics in a Larix gmelinii forest ecosystem[J]. Journal of Beijing Forestry University, 2015, 37(8): 31-39. DOI: 10.13332/j.1000-1522.20150020
[8]	LI Xin-yu, LI Yan-ming, SUN Lin, XU Rui, ZHAO Song-ting, GUO Jia. Characteristics of transpiration water consumption and its relationship with environmental factors in Ginkgo biloba[J]. Journal of Beijing Forestry University, 2014, 36(4): 23-29. DOI: 10.13332/j.cnki.jbfu.2014.04.008
[9]	ZHAO Bing-qing, WANG Yun-qi, WANG Bin, WANG Yu-jie, ZHANG Hui-lan. Role of environmental factors on forest soil respiration characteristics in Jinyun Mountain of Chongqing, southwestern China[J]. Journal of Beijing Forestry University, 2014, 36(3): 83-89. DOI: 10.13332/j.cnki.jbfu.2014.03.012
[10]	TIAN Jing-hui, HE Kang-ning, WANG Bai-tian, ZHANG Wei-qiang, YIN Jing. Relationship between transpiration of Platycladus orientalis and environmental factors in semi-arid region on Loess Plateau[J]. Journal of Beijing Forestry University, 2005, 27(3): 53-56.

Cited By

Cited by

Periodical cited type(4)

1.	李树萍，董琼，李世民，金友帆，张梅. 树番茄幼苗生长及氮积累与分配对光照和氮素添加的响应. 江西农业大学学报. 2023(01): 156-168 .
2.	刘博达. 氮肥对油松幼苗生长的影响. 特种经济动植物. 2023(11): 28-30 .
3.	潘陆荣，周袁慧子，王艺锦，苏远玉，王凌晖. 遮光和施肥对苹婆幼苗生理特性和氮素积累的影响. 湖北农业科学. 2021(14): 79-85 .
4.	李惋瑾，王若水，肖辉杰，王百田，张克斌，刘青青，郭冰寒. 鲜海带生物酶解有机液肥对沙木蓼生长和土壤理化性质的影响. 北京林业大学学报. 2018(07): 62-72 . 本站查看

Other cited types(12)

Get Citation

PDF

XML

Article views (1862) PDF downloads (108) Cited by(16)

Turn off MathJax

Article Contents

Abstract

References

Forest fire identification based on Sparse-DenseNet model