Successful remote sensing image registration is an important step for many remote sensing applications. The scale-invariant feature transform (SIFT) is a well-known method for remote sensing image registration, with many variants of SIFT proposed. However, it only uses local low-level information, and loses much middle- or high-level information to register. Image features extracted by a convolutional neural network (CNN) have achieved the state-of-the-art performance for image classification and retrieval problems, and can provide much middle- and high-level information for remote sensing image registration. Hence, in this letter, we investigate how to calculate the CNN feature, and study the way to fuse SIFT and CNN features for remote sensing image registration. The experimental results demonstrate that the proposed method yields a better registration performance in terms of both the aligning accuracy and the number of correct correspondences.

Various investigations have shown that driver fatigue is the main cause of traffic accidents. Research on the use of computer vision techniques to detect signs of fatigue from facial actions, such as yawning, has demonstrated good potential. However, accurate and robust detection of yawning is difficult because of the complicated facial actions and expressions of drivers in the real driving environment. Several facial actions and expressions have the same mouth deformation as yawning. Thus, a novel approach to detecting yawning based on subtle facial action recognition is proposed in this study to alleviate the abovementioned problems. A 3D deep learning network with a low time sampling characteristic is proposed for subtle facial action recognition. This network uses 3D convolutional and bidirectional long short-term memory networks for spatiotemporal feature extraction and adopts SoftMax for classification. A keyframe selection algorithm is designed to select the most representative frame sequence from subtle facial actions. This algorithm rapidly eliminates redundant frames using image histograms with low computation cost and detects outliers by median absolute deviation. A series of experiments are also conducted on YawDD benchmark and self-collected datasets. Compared with several state-of-the-art methods, the proposed method has high yawning detection rates and can effectively distinguish yawning from similar facial actions.

Automatic human fall detection is one important research topic in caring for vulnerable people such as elders at home and patients in medical places. Over the past decade, numerous methods aiming at solving the problem were proposed. However, the existing methods only focus on detecting human themselves and cannot work effectively in complicated environments, especially for the falls on furniture. To alleviate this problem, a new method for human fall detection on furniture using scene analysis based on deep learning and activity characteristics is presented in this paper. The proposed method first performs scene analysis using a deep learning method Faster R-CNN to detect human and furniture. Meanwhile, the space relation between human and furniture is detected. The activity characteristics of the detected people such as human shape aspect ratio, centroid, motion speed are detected and tracked. Through measuring the changes of these characteristics and judging the relations between people and furniture nearby, the falls on furniture can be effectively detected. Experiment results demonstrated that our approach not only accurately and effectively detected falls on furniture such as sofa and chairs but also distinguished them from other fall-like activities such as sitting or lying down, while the existing methods have difficulties to handle these. In our experiments, our algorithm achieved 94.44% precision, 94.95% recall and 95.50% accuracy. The proposed method can be potentially used and integrated as a medical assistance in health care and medical places and appliances.

Traffic sign recognition (TSR) plays an important role in driving assistance system and traffic safety insurance. However, existing methods focus on extracting features of traffic signs and ignore the constraints of spatial positional relationships between traffic signs and other objects in the scene. This way results in incorrectly detecting other similar objects as traffic signs and failing to detect very small traffic signs. A TSR method based on semantic scene understanding and structural traffic sign location is proposed in this study to solve the aforementioned problems. A scene structure model based on the constraints of spatial positional relationships between traffic signs and other objects is proposed to establish trusted search regions. An improved Light-weight RefineNet is used to analyze and understand a scene semantically and accurately and then segment objects in complicated environments precisely. A new network multiscale densely connected object detector (MDCOD) based on densely connected style, multiscale feature fusion, and improved K-means++ algorithms is proposed to recognize very small traffic signs. The trusted traffic signs are found by filtering false candidates outside the scene structure model. The proposed method is tested on Tsinghua-Tencent 100K and German Traffic Sign Detection Benchmark datasets and achieves accuracies of 92.8% and 99.90%, respectively, outperforming the existing methods.

Remote sensing image retrieval (RSIR) is a fundamental task in remote sensing. Most content-based RSIR approaches take a simple distance as similarity criteria. A retrieval method based on weighted distance and basic features of convolutional neural network (CNN) is proposed in this letter. The method contains two stages. First, in offline stage, the pretrained CNN is fine-tuned by some labeled images from the target data set, then used to extract CNN features, and labeled the images in the retrieval data set. Second, in online stage, we use the fine-tuned CNN model to extract the CNN feature of the query image and calculate the weight of each image class and apply them to calculate the distance between the query image and the retrieved images. Experiments are conducted on two RSIR data sets. Compared with the state-of-the-art methods, the proposed method is simplified but efficient, significantly improving retrieval performance.