• Korean Journal of Remote Sensing
• /
• v.14 no.4
• /
• pp.343-352
• /
• 1998

Spectral reflectance data of upland crops at OSMI bands were collected and evaluated for the feasibility of crop discrimination knowledge-based on crop calendar. Effective bands and their ratio values for discriminating corn from two other legumes were defined with OSMI equivalent bands and their ratio values. For corn discrimination from two other legumes, peanut and soybean, June 22 among measurements dates was the best since all OSMI equivalent bands and their ratio values in June 22 were highly significant for corn separability. Phenological growth stage of a silage corn (rs510) could be estimated as a function of spectral reflectance in vegetative stage. Five growth stage prediction models were generated by the SAS procedures REG and STEPWISE with OSMI equivalent bands and their ratio values in vegetative stage.

• Proceedings of the KSRS Conference
• /
• 1998.09a
• /
• pp.370-375
• /
• 1998

Spectral reflectance signatures of upland crops at OSMI bands were collected and evaluated for the feasibility of crop discrimination knowledge-based on crop calendar. Effective bands and their ratio values for discriminating corn from two other legumes were defined with OSMI equivalent bands and their ratio values. June 22 among measurements dates was the best date for corn discrimination from two other legumes, peanut and soybean, because all OSMI equivalent bands and their ratio values in June 22 were highly significant for corn separability. Phenological growth stage of a silage corn (rs510) could be estimated as a function of spectral reflectance signatures in vegetative stage. Five growth stage prediction models were generated by the SAS procedures REG and STEPWISE with OSMI equivalent bands and their ratio values in vegetative stage.

• Korean Journal of Remote Sensing
• /
• v.37 no.4
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.