• Laser & Optoelectronics Progress
  • Vol. 56, Issue 2, 021702 (2019)
Miao Yan1,2, Hongdong Zhao1,*, Yuhai Li2, Jie Zhang1,2, and Zetong Zhao1,2
Author Affiliations
  • 1 School of Electronics and Information Engineering, Hebei University of Technology, Tianjin 300401, China
  • 2 Electronics Technology Group Corporation No.53 Research Institute, Key Laboratory of Electro-Optical Information Control and Security Technology, Tianjin 300308, China
  • show less
    DOI: 10.3788/LOP56.021702 Cite this Article Set citation alerts
    Miao Yan, Hongdong Zhao, Yuhai Li, Jie Zhang, Zetong Zhao. Multi-Classification and Recognition of Hyperspectral Remote Sensing Objects Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(2): 021702 Copy Citation Text show less
    Flow chart of proposed remote sensing image classification
    Fig. 1. Flow chart of proposed remote sensing image classification
    Image examples of remote sensing ground objects
    Fig. 2. Image examples of remote sensing ground objects
    Learning rate comparison of neural network models with different pooling layers and classifiers. (a) Max pooling, Softmax classifier; (b) Max pooling, Sigmoid classifier; (c) Mean pooling, Softmax classifier; (d) Mean pooling, Sigmoid classifier
    Fig. 3. Learning rate comparison of neural network models with different pooling layers and classifiers. (a) Max pooling, Softmax classifier; (b) Max pooling, Sigmoid classifier; (c) Mean pooling, Softmax classifier; (d) Mean pooling, Sigmoid classifier
    Recognition rate comparison of four models for different iteration times
    Fig. 4. Recognition rate comparison of four models for different iteration times
    Accuracy comparison of three datasets under optimal parameters
    Fig. 5. Accuracy comparison of three datasets under optimal parameters
    ModelModelⅠModelⅡModelⅢModelⅣ
    NetworkmodelConvl (6@5×5)Convl (6@5×5)Convl (6@5×5)Convl (6@5×5)
    Max pooling (2×2)Max pooling (2×2)Mean pooling (2×2)Mean pooling (2×2)
    Convl (12@5×5)Convl (12@5×5)Convl (12@5×5)Convl (12@5×5)
    Max pooling (2×2)Max pooling (2×2)Mean pooling (2×2)Mean pooling (2×2)
    Softmax classifierSigmoid classifierSoftmax classifierSigmoid classifier
    Optimal learn rate0.030.30.60.9
    Accuracy0.96580.95830.96040.91
    Table 1. Comparison of neural network models
    Iteration timesABCDEFTotal
    20018005440238
    4009805000148
    60010550056
    80000680068
    100000630467
    120000360036
    140000330033
    160010280029
    180000360036
    200000350035
    220000360036
    Table 2. Number of unclassified images in test set for dataset-I
    Iteration timesABCDEFGHIJTotal
    20019104120540111168567
    40017063102208588276
    60015036101001162135
    800140562080531116
    10005039204013384
    120010631040834111
    140000611060102098
    16000050202062686
    18004058101012994
    200000491010112587
    22002045101022172
    Table 3. Number of unclassified images in test set for dataset-Ⅱ
    Iteration timesABCDEFGHIJKLMNTotal
    200160029440780168183116675424923
    4002502550270169063181410293
    6000091493407538451138318
    800250334080763301086194
    1000004400160194037376172
    1200720833870047341973238
    1400005620160113231475164
    160000581050123630555157
    18000010600120682831355258
    20003030009003432465123
    2200005700150162430455156
    Table 4. Number of unclassified images in test set for dataset-Ⅲ
    Miao Yan, Hongdong Zhao, Yuhai Li, Jie Zhang, Zetong Zhao. Multi-Classification and Recognition of Hyperspectral Remote Sensing Objects Based on Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(2): 021702
    Download Citation