No abstract

In Vietnam, prohibitions on using the telephone while driving vehicles have effected in a long time. However, that is relatively common and dangerous for traffic participants in reality. Along with the continuous growing of smartphones, using phone while driving is becoming a bad habit and one of the major causes of crashes and accidents of traffic in Vietnam. MotorSafe is a novel application that has been practically implemented as a solution for this problem. From the obtained data of the accelerometer sensor on the smartphone, a proposed decision tree algorithm enables device to recognize the user's  status on driving a vehicle. Some necessary tasks are provided that will help drivers focus on driving instead of unnecessary behaviors.

<abstract> <p>Monitor and classify behavioral activities in cows is a helpful support solution for livestock based on the analysis of data from sensors attached to the animal. Accelerometers are particularly suited for monitoring cow behaviors due to small size, lightweight and high accuracy. Nevertheless, the interpretation of the data collected by such sensors when characterizing the type of behaviors still brings major challenges to developers, related to activity complexity (i.e., certain behaviors contain similar gestures). This paper presents a new design of cows' behavior classifier based on acceleration data and proposed feature set. Analysis of cow acceleration data is used to extract features for classification using machine learning algorithms. We found that with 5 features (mean, standard deviation, root mean square, median, range) and 16-second window of data (1 sample/second), classification of seven cow behaviors (including feeding, lying, standing, lying down, standing up, normal walking, active walking) achieved the overall highest performance. We validated the results with acceleration data from a public source. Performance of our proposed classifier was evaluated and compared to existing ones in terms of the sensitivity, the accuracy, the positive predictive value, and the negative predictive value.</p> </abstract>

No abstract

The daily behavior of dairy cows reflects the health status and well being. An automated monitoring system is needed for suitable management. It helps farmers to have a comprehensive view of the cattle healthy and manage large of cows. Acceleration sensors can be found in various kinds of applications. In this paper, we detect the cow’s activities by using a multidimensional acceleration sensor and multiclass support vector machine (SVM). The acceleration sensor is attached to the cow’s neck-collar in order to sense the movements in X, Y, and Z axes. The data is brought to a microprocessor for pre-processing, and join in a wireless sensor network (WSN) through a Zigbee module. After that, the data are transferred to the server. At the server, a suitable SVM algorithm is chosen and applied to classify four main behaviors: standing, lying, feeding and walking. A well know kernels, Radius Basic Function (RBF), is chosen. After that, a cross validation (k-fold) is used to measure the error and select the best fit model. The sensor is used to acquire experimental data from Vietnam Yellow cows in the cattle farm. The promising results with the average sensitivity of 87.51%, and the average precision of 90.24% confirm the reliability of our solution. The classification results can be automatically uploaded to the cloud internet and the farmer can easily access to check the status of his cows.