Research on Real-Time Face Key Point Detection Algorithm Based on Attention Mechanism

The existing face detection methods were affected by the network model structure used. Most of the face recognition methods had low recognition rate of face key point features due to many parameters and large amount of calculation. In order to improve the recognition accuracy and detection speed of face key points, a real-time face key point detection algorithm based on attention mechanism was proposed in this paper. Due to the multiscale characteristics of face key point features, the deep convolution network model was adopted, the attention module was added to the VGG network structure, the feature enhancement module and feature fusion module were combined to improve the shallow feature representation ability of VGG, and the cascade attention mechanism was used to improve the deep feature representation ability. Experiments showed that the proposed algorithm not only can effectively realize face key point recognition but also has better recognition accuracy and detection speed than other similar methods. This method can provide some theoretical basis and technical support for face detection in complex environment.

Array model of shock pressure sensor for shooting point detection

The array model of double-T shock pressure sensor is established. Shock wave is produced by a supersonic object in the air. Pressure is produced in the process of shock wave transmission. Different shock pressure sensors have different time to receive the pressure signal. In this paper, the shooting point calculation model and the finite element model of the double T-shaped array method are established. The simulation experiment is carried out. The law of shock wave propagation is verified. The model can be used to calculate the coordinates of shooting point quickly. This method is suitable for small angle oblique fire location problem, and improves the detection accuracy of shooting point.

A Deep Learning Based System for the Detection of Human Violence in Video Data

The number of security cameras positioned within the surrounding area has expanded, increasing the demand for automatic activity recognition systems. In addition to offline assessment and the issuance of an ongoing alarm in the case of aberrant behaviour, automatic activity detection systems can be employed in conjunction with human operators. In the proposed research framework, an ensemble of Mask Region-based Convolutional Neural Networks for key-point detection scheme, and LSTM based Recurrent Neural Network is used to create a deep neural network model (Mask RCNN) for recognizing violent activities (i.e. kicking, punching, etc.) of a single person. First of all, the key-points locations and ground-truth masks of humans in an image are selected using the selected region; the temporal information is extracted. Experimental results show that the ensemble model outperforms individual models. The proposed technique has a reasonable accuracy rate of 77.4 percent, 95.7 percent, and 88.2 percent, respectively, on the Weizmann, KTH, and our custom datasets. As the proposed effort applies to industry and in terms of security, it is beneficial to society.

Vision-based vanishing point detection of autonomous navigation of mobile robot for outdoor applications

The vision-based approach to mobile robot navigation is considered superior due to its affordability. This paper aims to design and construct an autonomous mobile robot with a vision-based system for outdoor navigation. This robot receives inputs from camera and ultrasonic sensor. The camera is used to detect vanishing points and obstacles from the road. The vanishing point is used to detect the heading of the road. Lines are extracted from the environment using a canny edge detector and Houghline Transforms from OpenCV to navigate the system. Then, removed lines are processed to locate the vanishing point and the road angle. A low pass filter is then applied to detect a vanishing point better. The robot is tested to run in several outdoor conditions such as asphalt roads and pedestrian roads to follow the detected vanishing point. By implementing a Simple Blob Detector from OpenCV and ultrasonic sensor module, the obstacle's position in front of the robot is detected. The test results show that the robot can avoid obstacles while following the heading of the road in outdoor environments. Vision-based vanishing point detection is successfully applied for outdoor applications of autonomous mobile robot navigation.