深度学习在植物叶部病害检测与识别的研究进展
邵明月, 张建华, 冯全, 柴秀娟, 张凝, 张文蓉

Research Progress of Deep Learning in Detection and Recognition of Plant Leaf Diseases
SHAO Mingyue, ZHANG Jianhua, FENG Quan, CHAI Xiujuan, ZHANG Ning, ZHANG Wenrong
表5 近年来基于深度网络的植物病害识别研究进展
Table 5 Recent advances in plant disease recognition based on deep network
编号作者年份植物种类数据集/幅获取方法神经网络类型最高准确率/%
1Jiang等862021

水稻

小麦

水稻:120

小麦:80

田间拍摄VGG-16

水稻:97.22

小麦:98.75

2Abbas等872021番茄16,012PlantVillageDenseNet12199.51
3Chellapandi等88202114种植物54,306PlantVillageDenseNet99.00
4樊湘鹏等892021葡萄19,900田间拍摄改进VGG1698.02
5Jiang等902020水稻8911田间拍摄CNN+SVM96.80
6Barman等912020柑橘2939田间拍摄SSCNN98.00
7Dang等922020萝卜40可控环境下拍摄GoogLeNet90.00
8Ji等782020葡萄1619PlantVillageUnitedModel99.17
9Howlader等932019石榴2705田间拍摄D-CNN98.74
10Coulibaly等942019粟米124田间拍摄VGG1695.00
11Hu等952019144田间拍摄CIFAR-10Net92.50
12Sibiya和Sumbwanyambe962019玉米100田间拍摄50层CNN92.85
13王艳玲等972019番茄14,529PlantVillageAlexNet95.62
14Singh等982019芒果1070田间拍摄MCNN97.13
15Picon等992019小麦8178田间拍摄改进ResNet-5096.00
16Abdalla等1002019油菜籽400田间拍摄VGG1696.00
17Xing等762019柑橘12,561田间拍摄Weakly DenseNet93.42
18曾伟辉等77201947种植物56,190MK- D2、 PlantVillage、 AES- CD9214HORPSF96.75
19Atole和Park1012018水稻227田间拍摄AlexNet91.23
20Zhang等1022018玉米500PlantVillage+线上采集GoogLeNet98.90
21Liu等1032018苹果13,689可控环境下拍摄AlexNet97.62
22张建华等1042018棉花5510田间拍摄VGG-1689.51
23Rangarajan等1052018番茄13,262PlantVillageVGG1697.49
24Ferentinos74201825种植物87,00037.3%田间拍摄,62.7%可控条件拍摄VGG99.53
25赵建敏等752018马铃薯6000田间拍摄8层CNN87.00
26De Chant等1062017玉米1796田间拍摄CNN96.70
27Lu等1072017水稻500田间拍摄CNN95.00
28Oppenheim和Shani1082017马铃薯2465可控环境下拍摄VGG90.00
29Ramcharan等712017木薯2756田间拍摄Inception v393.00
30孙俊等72201714种植物21,917PlantVillage改进的AlexNet99.41
31Lu等732017小麦9230田间拍摄VGG-FCN-VD1697.95
32Fujita等1092016黄瓜7520田间拍摄CNN82.30
33Sladojevic等69201613种植物2589线上采集Caffe+迁移学习96.30
34Mohanty等70201614种植物54,306PlantVillageAlexNet99.35
35Kawasaki等682015黄瓜800田间拍摄CNN94.90