In this paper, we explore the elicitation of age and gender information from gait traces obtained from IMU sensors, and systematically compare different feature engineering and machine learning algorithms, including both traditional and deep learning methods. Far worse, the leakage of sensitive information to third parties, such as recommender systems or targeted advertising applications, may cause privacy concerns for unsuspecting end-users. However, these sensors may also incidentally reveal sensitive information in a way that is not easily envisioned upfront by developers. They have been widely adopted in the area of activity recognition, with fall detection and step counting applications being prominent examples in this field. Inertial measurement unit (IMU) sensors-and accelerometers and gyroscopes in particular-are readily available on contemporary smartphones and wearable devices.
Sensors provide the foundation of many smart applications and cyber–physical systems by measuring and processing information upon which applications can make intelligent decisions or inform their users. The results show that FPRF-GR can identify five gaits (walk, stationary, run, and up and down stairs) with the average accuracy of 98.2%. Finally, this paper proposes an optimization scheme for the two parameters of decision tree number and sample number in RF. FPRF-GR builds the model based on RF, and uses the tenfold cross validation method to evaluate the model. Then, in the design of classifier, in order to meet the requirements of gait recognition model for accuracy, generalization ability, speed, and noise resistance, this paper compares random forest (RF) and several commonly used classification algorithms, and finds that the model constructed by RF can meet the requirements. Firstly, a fusion feature engineering operator is designed to eliminate redundant and defective features, which is mainly based on Fast Fourier Transform and principal component analysis. Therefore, this paper proposes a gait recognition algorithm based on IMU, which is named as FPRF-GR. With the rapid development of micro-electro mechanical systems, inertial measurement unit (IMU) has been widely used in the field of gait recognition with many advantages, such as low cost, small size, and light weight. In recent years, gait detection has been widely used in medical rehabilitation, smart phone, criminal investigation, navigation and positioning and other fields. We open-source all contributions for re-producibility and broader use by the research community. Our proposed approach can serve as a baseline utility to evaluate the privacy-invasiveness of artificial intelligence systems that make use of privacy-sensitive information. In our experiments we were able to accurately predict sex (94.7%),race (83.1%) and age (MAE 6.21 and RMSE 8.33) from masked face images. Despite the popular belief of the privacy benefits of wearing a mask among people, we show that there is no significant difference to privacy invasiveness when a mask is worn. We train and apply a CNN based on the ResNet-50 architecture with 20,003 synthetic masked images and measure the privacy invasiveness. In this work, we analyse the privacy invasiveness of face biometric systems by predicting privacy-sensitive soft-biometrics using masked face images. Despite the positive benefits of such systems, there is potential for exploitation by invading user privacy. Contactless and efficient systems are implemented rapidly to advocate preventive methods in the fight against the COVID-19 pandemic.