Impact of shooting models on four moth species identification by using automatic recognition system in Fujian province

doi:10.14088/j.cnki.issn0439-8114.2019.08.032

HUBEI AGRICULTURAL SCIENCES ›› 2019, Vol. 58 ›› Issue (8): 136-140.doi: 10.14088/j.cnki.issn0439-8114.2019.08.032

• Agricultural Engineering • Previous Articles Next Articles

Impact of shooting models on four moth species identification by using automatic recognition system in Fujian province

LU Ci-ding^a, HUA Yin^a, LIN Da-kun^b, HAN Xiao-hong^a, LIN Hao-yu^a, CAI Xiao-ming^a, HUANG Shi-guo^b, ZHANG Fei-ping^a, LIANG Guang-hong^a

a.Forestry College; b. College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou 350002, China

Received:2018-12-27 Online:2019-04-25 Published:2019-12-04

Abstract

Abstract: This study was performed to figure out the impact of shooting models on moth recognition based on images of hundred of specimens from four common forest moths in Fujian province. All specimens images were taken and collected under different shooting mode, and recognized and analyzed by using MATLAB 2009 and statistical software. The results showed that： ① The shutter speed of the SLR camera had a significant impact on the recognition rate （P<0.05）, with 1/2000s or 1/4000s as the best shutter speed, however, the sensitivity （ISO） and aperture （f） size had no significant influence on the recognition rate （P>0.05）. ② The recognition rate of SLR （93.3%） camera was the highest among three cameras, but there were not remarkable impact of different three cameras on the recognition rate （P>0.05）. ③ There was a significant influence on recognition rate among different shooting angles （P<0.05）, and the angles range from 90°to 120° were the best. Consequently, to improve the image quality and the recognition rate, we suggested operator try to use shooting angles from 90°to 120°, combining the shutter speed and angles together. This research is supposed to lay a foundation for the optimization, application and promotion of moths image recognition technology in the future.

Key words: specimen, moths, automatic recognition system, recognition rate

CLC Number:

S763
TP399

LU Ci-ding, HUA Yin, LIN Da-kun, HAN Xiao-hong, LIN Hao-yu, CAI Xiao-ming, HUANG Shi-guo, ZHANG Fei-ping, LIANG Guang-hong. Impact of shooting models on four moth species identification by using automatic recognition system in Fujian province[J]. HUBEI AGRICULTURAL SCIENCES, 2019, 58(8): 136-140.

References

[1] 陈顺立,林庆源,黄金聪.南方主要树种害虫综合管理[M].福建厦门:厦门大学出版社,2003.
[2] 丁莉萍. 福建园林植物病虫害防治[M].北京:中国林业出版社,2014.
[3] 王江宁,纪力强.昆虫图像特征研究[J].山东大学学报(工学版),2011,41(2):51-57.
[4] 南宫自艳,李静,白向宾.昆虫分类学主要技术手段的研究进展[J].环境昆虫学报,2014,36(6):1004-1010.
[5] WEEKS P J D,O'NEILL M A,GASTON K J,et al. Automating insect identification:Exploring the limitations of a prototype system[J].Journal of applied entomology,2003,123(1):1-8.
[6] RATNIEKS F L W,SCHRELL F,REBECCA C S,et al. Data reliability in citizen science:Learning curve and the effects of training method,volunteer background and experience on identification accuracy of insects visiting ivy flowers[J].Methods in ecology and evolution,2016,7(10):1226-1235.
[7] 竺乐庆,张大兴,张真.基于韦伯局部描述子和颜色直方图的鳞翅目昆虫翅图像特征描述与种类识别[J].昆虫学报,2015, 58(4):419-426.
[8] 李小林,周蓬勃,周明全,等.基于可区分二进制局部模式特征的蛾类昆虫识别[J].计算机应用与软件,2016,33(3):172-175.
[9] 刘杰,姜玉英,赵友文,等.昆虫标本的采集处理和微距摄影技术[J].农业工程,2015,5(4):146-149.
[10] 蔡小娜,黄大庄,沈佐锐,等.蛾翅翅脉特征在夜蛾昆虫数学分类学中的应用[J].中国农业大学学报,2015,20(5):188-193.
[11] 黄世国. 基于图像的昆虫识别关键技术研究[D].西安:西北大学,2008.
[12] 董学超. 数学形态特征在夜蛾科昆虫分类鉴定上的应用研究[D].河北保定:河北农业大学,2014.
[13] 苏筱雨,蔡小娜,李阳,等.利用翅的数学形态特征对8种金龟子进行分类鉴定[J].中国农业大学学报,2016,21(7):61-66.
[14] 李阳,沈佐锐,董学超,等.利用翅的数学形态特征对20种夜蛾进行分类鉴定的研究[J].中国农业大学学报,2016,21(9):80-89.

Impact of shooting models on four moth species identification by using automatic recognition system in Fujian province

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 1

Recommended Articles

Metrics

Comments