In recent years, researchers of deep neural networks (DNNs)-based facial expression recognition (FER) have reported results showing that these approaches overcome the limitations of conventional machine learning-based FER approaches. However, as DNN-based FER approaches require an excessive amount of memory and incur high processing costs, their application in various fields is very limited and depends on the hardware specifications. In this paper, we propose a fast FER algorithm for monitoring a driver’s emotions that is capable of operating in low specification devices installed in vehicles. For this purpose, a hierarchical weighted random forest (WRF) classifier that is trained based on the similarity of sample data, in order to improve its accuracy, is employed. In the first step, facial landmarks are detected from input images and geometric features are extracted, considering the spatial position between landmarks. These feature vectors are then implemented in the proposed hierarchical WRF classifier to classify facial expressions. Our method was evaluated experimentally using three databases, extended Cohn-Kanade database (CK+), MMI and the Keimyung University Facial Expression of Drivers (KMU-FED) database, and its performance was compared with that of state-of-the-art methods. The results show that our proposed method yields a performance similar to that of deep learning FER methods as 92.6% for CK+ and 76.7% for MMI, with a significantly reduced processing cost approximately 3731 times less than that of the DNN method. These results confirm that the proposed method is optimized for real-time embedded applications having limited computing resources.

This study proposes a lightweight multilayer random forest (LMRF) model, which is a non-neural network style deep model consisting of layer-by-layer random forests. Although a deep neural network (DNN) is a powerful algorithm for facial expression recognition (FER), the requirement of too many parameters, careful parameter tuning, a huge amount of training data, black-box models, and a pre-trained architecture remain significant burdens for a current DNN, particularly for real-time processing. To overcome the limitations of a DNN, our FER system uses LMRF consisting of a two-layer structure and a small number of trees per layer for fast FER. The proposed LMRF achieves a similar performance as a DNN even with a small number of hyper-parameters, and a faster processing time using a CPU. We conducted experiments using a benchmark database captured indoors and a real driving database captured using a near-infrared (NIR) camera. Based on a performance evaluation against a few other state-of-the-art FER methods, the proposed method showed a more uniform performance than DNN-based methods, and required a reduced number of parameters and operations without a loss of accuracy when compared to DNN model compression algorithms. As a replacement for deeper and wider networks, the proposed model can be embedded in low-power and low-memory in vehicle systems for the monitoring of a driver's emotion. INDEX TERMS Non-neural networks, deep random forest, driver emotion monitoring, facial expression recognition, multilayer random forest.

No abstract

No abstract

As the need for wildfire detection increases, research on wildfire smoke detection combining low-cost cameras and deep learning technology is increasing. Camera-based wildfire smoke detection is inexpensive, allowing for a quick detection, and allows a smoke to be checked by the naked eye. However, because a surveillance system must rely only on visual characteristics, it often erroneously detects fog and clouds as smoke. In this study, a combination of a You-Only-Look-Once detector and a long short-term memory (LSTM) classifier is applied to improve the performance of wildfire smoke detection by reflecting on the spatial and temporal characteristics of wildfire smoke. However, because it is necessary to lighten the heavy LSTM model for real-time smoke detection, in this paper, we propose a new method for applying the teacher–student framework to deep LSTM. Through this method, a shallow student LSTM is designed to reduce the number of layers and cells constituting the LSTM model while maintaining the original deep LSTM performance. As the experimental results indicate, our proposed method achieves up to an 8.4-fold decrease in the number of parameters and a faster processing time than the teacher LSTM while maintaining a similar detection performance as deep LSTM using several state-of-the-art methods on a wildfire benchmark dataset.