• Korean Journal of Agricultural and Forest Meteorology
• /
• v.26 no.2
• /
• pp.89-102
• /
• 2024

Land Surface Phenology (LSP) plays a crucial role in understanding vegetation dynamics. The near-infrared reflectance of vegetation (NIRv) has been increasingly adopted in LSP studies, being recognized as a robust proxy for gross primary production (GPP). However, NIR v is sensitive to the terrain effects in mountainous areas due to artifacts in NIR reflectance cannot be canceled out. Because of this, estimating phenological metrics in mountainous regions have a substantial uncertainty, especially in the end of season (EOS). The topographically corrected NIRv (TCNIRv) employs the path length correction (PLC) method, which was deduced from the simplification of the radiative transfer equation, to alleviate limitations related to the terrain effects. TCNIRv has been demonstrated to estimate phenology metrics more accurately than NIRv, especially exhibiting improved estimation of EOS. As the topographic effect is significantly influenced by terrain properties such as slope and aspect, our study compared phenology metrics estimations between south-facing slopes (SFS) and north-facing slopes (NFS) using NIRv and TCNIRv in two distinct mountainous regions: Gwangneung Forest (GF) and Odaesan National Park (ONP), representing relatively flat and rugged areas, respectively. The results indicated that TCNIR v-derived EOS at NFS occurred later than that at SFS for both study sites (GF : DOY 266.8/268.3 at SFS/NFS; ONP : DOY 262.0/264.8 at SFS/NFS), in contrast to the results obtained with NIRv (GF : DOY 270.3/265.5 at SFS/NFS; ONP : DOY 265.0/261.8 at SFS/NFS). Additionally, the gap between SFS and NFS diminished after topographic correction (GF : DOY 270.3/265.5 at SFS/NFS; ONP : DOY 265.0/261.8 at SFS/NFS). We conclude that TCNIRv exhibits discrepancy with NIR v in EOS detection considering slope orientation. Our findings underscore the necessity of topographic correction in estimating photosynthetic phenology, considering slope orientation, especially in diverse terrain conditions.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.