Search

For the advancement of smart grids, solar power generation predictions have become an important research topic. In the case of using traditional modeling methods, excessive computational costs may be incurred and it is difficult for these methods to learn the multi-variable dependencies of the data. Therefore, in this paper, a deep learning model was used to combine convolutional neural networks and long short-term memory recurrent network predictions. This method enables hourly power generation one day into the future. Convolutional neural networks are used to extract the features of multiple time series, while long short-term memory neural networks predict multivariate outcomes simultaneously. In order to obtain more accurate prediction results, we performed feature selection on meteorological features and combined the selected weather features to train the prediction model. We further distinguished sunny- and rainy-day models according to the predicted daily rainfall conditions. In the experiment, it was shown that the method of combining meteorological features further reduced the error. Finally, taking into account the differences in climate conditions between the northern and southern regions of Taiwan, the experimental results of case studies involving multiple regions were evaluated to verify the proposed method. The results showed that training combined with selected meteorological features can be widely used in regions with different climates in Taiwan. [ABSTRACT FROM AUTHOR]

Video action recognition is vital in the research area of computer vision. In this paper, we develop a novel model, named Separable ConvNet Spatiotemporal Mixer (SCSM). Our goal is to develop an efficient and lightweight action recognition backbone that can be applied to multi-task models to increase the accuracy and processing speed. The SCSM model uses a new hierarchical spatial compression, employing the spatiotemporal fusion method, consisting of a spatial domain and a temporal domain. The SCSM model maintains the independence of each frame in the spatial domain for feature extraction and fuses the spatiotemporal features in the temporal domain. The architecture can be adapted to different frame rate requirements due to its high scalability. It is suitable to serve as a backbone for multi-task video feature extraction or industrial applications with its low prediction and training costs. According to the experimental results, SCSM has a low number of parameters and low computational complexity, making it highly scalable with strong transfer learning capabilities. The model achieves video action recognition accuracy comparable to state-of-the-art models with a smaller parameter size and fewer computational requirements. [ABSTRACT FROM AUTHOR]

In this article, we propose a pose transfer framework that can deal with large body motion to generate dance videos. To solve the problem of body shape deformation from large movements, a multiple scale region extractor (MSRE) is proposed. The features of each body region can be extracted from multiple layers of the encoder according to the body key points and passed through shortcuts to the decoder to reduce the spatial information loss. We add a region style loss calculated by the style representations of the body regions to the loss function to improve the quality of the generated images. In addition, the concept of learnable region normalization is integrated into the proposed framework to prevent introducing undesired mean and variance shifts by the corrupted regions during normalization. The experiments have shown that the proposed system can significantly improve the pose generation results compared with existing methods, especially when there are large body movements in the dancing poses. [ABSTRACT FROM AUTHOR]

In this work, a day-ahead to week-ahead solar irradiance prediction mechanism based on convolutional Long Short-Term Memory (LSTM) model is proposed. The system takes hourly irradiance data from several days previous to the prediction day as the input. Then, features are extracted from the input data using one dimensional convolutional filters. The extracted features from different days are concatenated and serve as the input of the LSTM network. The output of the LSTM is further concatenated with selected original data to emphasize its importance and enhance the prediction results. Afterwards, a fully connected layer is used to produce the final prediction output. The proposed framework can be trained using a relatively small amount of training data within the duration of only two months. Therefore, it has the advantage of being applicable in the initial deployment phase when the amount of training data is limited. The proposed system has been validated using a highly challenging dataset collected in Taiwan with tropical and subtropical marine island climate. [Display omitted] • A convolutional long short-term memory framework is designed for solar irradiance prediction. • Weather observation and forecast information are used in the training and prediction process. • The proposed framework can be trained with relatively small amount of data. • The proposed convolutional LSTM framework outperforms related works. [ABSTRACT FROM AUTHOR]

In this article, we propose a system for automatically classifying and cleaning location images in large-scale image databases uploaded by arbitrary users. Detecting incorrect scenes uploaded by users and maintaining the correctness of the database through automatic data cleaning are essential because human inspection is not feasible for verifying massive amounts of data. In this study, we compared different feature extractors using deep convolutional neural networks trained using big data. We designed a multilevel extractor to improve feature extraction. Moreover, a detector based on multiresolution dissimilarity calculation was designed to overcome the issue of large intraclass distances and successfully identify incorrect scenes. The proposed system was validated using a highly challenging dataset with 138,000 images collected from Google Places. The experiments show that the multilevel extractor and the detector based on multiresolution dissimilarity calculation can improve the accuracy in identifying incorrect scenes and achieve satisfying data cleaning results. [ABSTRACT FROM AUTHOR]

This paper proposes a framework to automatically conduct cloud classification on all-sky images and perform short-term solar irradiance prediction according to the classification results. The all-sky images are divided into blocks to deal with the mixed cloud type conditions. Local texture patterns and statistical texture features are extracted from the image blocks for cloud classification. Different cloud types with various heights, thickness, and opacity have different impact on the variation of solar irradiance. Therefore, several regression models are trained to capture the characteristics of irradiance changes under different cloud types. The current classified cloud type is used to select a corresponding prediction model. Such design substantially increases the prediction accuracy. The experimental results verify the effectiveness of the proposed framework. Both the proposed cloud classification method and irradiance prediction mechanism outperform existing works. Adding local texture patterns in the feature vector enhance the classification performance. Compared with non-block based methods, the proposed block-based method could increase the classification rate by 5%–10%. Utilizing multiple prediction models according cloud types could lower both the mean absolute error and the root mean squared error on short-term irradiance prediction. [ABSTRACT FROM AUTHOR]

In this paper, we propose an intelligent highway surveillance system that performs self-diagnosis and detects conditions when the camera is seriously tampered by raindrops at night. The system also provides solutions to analyze the traffic flow under the challenging nighttime raindrop-tampered conditions. To deal with the challenging scenes, we extract effective features via salient region detection and block segmentation. The extracted features are used to train a support vector machine to achieve self-diagnosis. For traffic flow analysis, we use the extracted features in the region of interest and construct a regression model to get an estimated vehicle count for each frame. The vehicle counts in consecutive frames form a vehicle count sequence. We propose a mapping model to acquire the desired per-minute traffic flow from the vehicle count sequence. The model utilizes state transfer likelihoods and takes into account the length of the segmented vehicle count sequence. The experiments on highly challenging data sets have demonstrated that the proposed system can effectively estimate the traffic flow for raindrop-tampered highway surveillance cameras at night. [ABSTRACT FROM PUBLISHER]

Highlights: [•] Moving skin regions and shadows are detected. [•] Skin and shadow probabilities are integrated into the mixture models. [•] Learning rates are adjusted by the skin and shadow probabilities. [•] Mandarin Phonetic Symbol (MPS) combination recognition algorithm is designed. [•] Single fingertip virtual keyboard system is implemented. [ABSTRACT FROM AUTHOR]