• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.504-507
• /
• 2020

ReLU(Rectified Linear Unit) 함수는 제안된 이후로 대부분의 깊은 인공신경망 모델들에서 표준 활성함수로써 지배적으로 사용되었다. 이후에 ReLU 를 대체하기 위해 Leaky ReLU, Swish, Mish 활성함수가 제시되었는데, 이들은 영상 분류 과업에서 기존 ReLU 함수 보다 향상된 성능을 보였다. 따라서 초해상화(Super Resolution) 과업에서도 ReLU 를 다른 활성함수들로 대체하여 성능 향상을 얻을 수 있는지 실험해볼 필요성을 느꼈다. 본 연구에서는 초해상화 과업에서 안정적인 성능을 보이는 EDSR(Enhanced Deep Super-Resolution Network) 모델의 활성함수들을 변경하면서 성능을 비교하였다. 결과적으로 EDSR 의 활성함수를 변경하면서 진행한 실험에서 해상도를 2 배로 변환하는 경우, 기존 활성함수인 ReLU 가 실험에 사용된 다른 활성함수들 보다 비슷하거나 높은 성능을 보였다. 하지만 해상도를 4 배로 변환하는 경우에서는 Leaky ReLU 와 Swish 함수가 기존 ReLU 함수대비 다소 향상된 성능을 보임을 확인하였다. 구체적으로 Leaky ReLU 를 사용했을 때 기존 ReLU 보다 영상의 품질을 정량적으로 평가할 수 있는 PSNR 과 SSIM 평가지표가 평균 0.06%, 0.05%, Swish 를 사용했을 때는 평균 0.06%, 0.03%의 성능 향상을 확인할 수 있었다. 4 배의 해상도를 높이는 초해상화의 경우, Leaky ReLU 와 Swish 가 ReLU 대비 향상된 성능을 보였기 때문에 향후 연구에서는 다른 초해상화 모델에서도 성능 향상을 위해 활성함수를 Leaky ReLU 나 Swish 로 대체하는 비교실험을 수행하는 것도 필요하다고 판단된다.

• Journal of the Korean Society of Radiology
• /
• v.14 no.3
• /
• pp.211-220
• /
• 2020

High-quality computed tomography (CT) images enable precise lesion detection and accurate diagnosis. A lot of studies have been performed to improve CT image quality while reducing radiation dose. Recently, deep learning-based techniques for improving CT image quality have been developed and show superior performance compared to conventional techniques. In this study, a super-resolution convolutional neural network (SRCNN) model was used to improve the spatial resolution of CT images, and image quality according to the hyperparameters, which determine the performance of the SRCNN model, was evaluated in order to verify the effect of hyperparameters on the SRCNN model. Profile, structural similarity (SSIM), peak signal-to-noise ratio (PSNR), and full-width at half-maximum (FWHM) were measured to evaluate the performance of the SRCNN model. The results showed that the performance of the SRCNN model was improved with an increase of the numbers of epochs and training sets, and the learning rate needed to be optimized for obtaining acceptable image quality. Therefore, the SRCNN model with optimal hyperparameters is able to improve CT image quality.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.24 no.3
• /
• pp.164-178
• /
• 2022

Uncertainties in weather forecasts would affect the reliability of yield prediction using crop models. The objective of this study was to compare uncertainty in crop yield prediction caused by the use of the weather forecast data. Daily weather data were produced at 10 km spatial resolution using W eather Research and Forecasting (W RF) model. The nearest neighbor method was used to downscale these data at the resolution of 5 km (W RF5K). Parameter-elevation Regressions on Independent Slopes Model (PRISM) was also applied to the WRF data to produce the weather data at the same resolution. W RF5K and PRISM data were used as inputs to the CROPGRO-SOYBEAN model to predict crop yield. The uncertainties of the gridded data were analyzed using cumulative growing degree days (CGDD) and cumulative solar radiation (CSRAD) during the soybean growing seasons for the crop of interest. The degree of agreement (DOA) statistics including structural similarity index were determined for the crop model outputs. Our results indicated that the DOA statistics for CGDD were correlated with that for the maturity dates predicted using WRF5K and PRISM data. Yield forecasts had small values of the DOA statistics when large spatial disagreement occured between maturity dates predicted using WRF5K and PRISM. These results suggest that the spatial uncertainties in temperature data would affect the reliability of the phenology and, as a result, yield predictions at a greater degree than those in solar radiation data. This merits further studies to assess the uncertainties of crop yield forecasts using a wide range of crop calendars.