Application of Human Activity Recognition in Supermarkets with Human Pose Estimation
Downloads
Human Activity Recognition is the ability to interpret human body movements or movements through sensors and to determine human activities or actions. Most everyday human tasks can be simplified or automated if they can be identified through an activity management system. The introduction of human activities in supermarkets has many benefits in its development, such as Just Walk Out technologies such as Amazon Go and the prevention of theft of goods. Proposed research with the theme of human activity recognition that has a focus on supermarkets. The methods that will be used in this research are Human Pose Estimation and Random Forest Classifier with data collected from the internet. The designed application will first detect skeletons with Blazepose from human objects, which will then be classified by the Random Forest Classifier. Activities that can be classified by the designed application are standing, walking, and fetching. The output of the application is a classification of activities carried out in real-time with the camera. The test results on the training data get an accuracy value of 100%, precision of 100%, recall of 100%, and F1-Score of 100%. Using the test data produces a confusion matrix which shows that the model being trained has an accuracy value of 100%, precision of 100%, recall of 100%, and an F1-Score of 100%.
Prastika, K., 2020, Aplikasi Pendeteksi Aktivitas Individu dalam Ruangan Menggunakan Metode CNN AlexNet, Computatio: Journal of Computer Science and Information Systems.
Hussain, Z., Sheng, M., dan Zhang, W. E., 2019, Different Approaches for Human Activity Recognition – A Survey, Journal of Network and Computer Applications, No. 102738, Vol. 167.
Gruosso, M., Capece, N., dan Erra, U., 2021, Human segmentation in surveillance video with deep learning, Multimedia Tools and Applications, Vol. 80, pp. 1175-1199.
Gupta, A., Gupta, K., Gupta, K., Gupta, K. Human Activity Recognition Using Pose Estimation and Machine Learning Algorithm, CEUR Workshop Proceedings, Vol. 2786.
Bazarevsky, V., Grishchenko, I., Raveendran, K., Zhu, T., Zhang, F., dan Grundmann, M., 2020, BlazePose: On-device Real-time Body Pose tracking, https://arxiv.org/abs/2006.10204, Last Accessed 1 September 2021.
Dogru, N., dan Subasi, A. Traffic accident detection using random forest classifier. In 15th learning and technology conference (L&T), pp. 40-45. IEEE.