Toktam nojavan - Computer Engineering Department Islamic Azad University, ferdows Branch, Iran
Hooman Kashanian - Assistant professor, Computer Engineering Department Islamic Azad University, ferdows Branch, Iran

Human activity Recognition is one of the topics has been the interest Researchers in recent decades. HAR is used for issues such as monitoring, performance evaluation and recognition of abnormal and suspicious activity, etc. Smart home environment is where data collected from sensors which installed in human body or environment to classify human activities. Appropriate diagnosis of daily life activities proposed to implement many strategies to encourage healthy behaviors related to diet, exercise and adherence to treatment will be necessary. Thus, in this paper a method is provided for Human activity Recognition in the smart home environment. This approach combination of neural networks is reversible and deep convolution. The proposed method implemented on the dataset which collected signals in the smart home. The results confirm the validity of the proposed method is in the 53 percent classified.

Human activity Recognition, neural networks, deep convolution, smart home