Pendeteksian Citra Pengunjung Menggunakan Single Shot Detector untuk Analisis dan Prediksi Seasonality
Main Article Content
This study discusses the analysis of retail store with time series method to obtain information about sales trend and seasonality by looking at visitor data and total transaction data at a time period. Data in the form of the number of customers who visit are obtained through CCTV video camera recordings placed at retail store X and the total transaction occurred at retail store X. The visitor counting uses the deep learning method with SSD (Single Shot Detector) object detection framework and MobileNet architecture. The library used to count the number of customers visiting the store is OpenCV, Pandas, Numpy, Dlib, and Imutils. The number of customers visiting the store will then be compared to the number of transactions that occur at the same time so that a conversion rate can be obtained. From here, we can see sales trend that occur at any time. Time series analysis is also carried out to determine and analyze the pattern of data obtained based on certain time to predict the things that need to be done in the future. Through this research, information has been successfully obtained related to seasonality patterns, value and interpretation of retail conversion rates, models for predicting the number of visitors and transactions, and answering the hypothesis with the Wilcoxon test method obtained a p-value of 0,014 which states that the data pattern of the number of consumers is not the same as the transaction data pattern.
Download data is not yet available.
How to Cite
D. Diamanta and H. Toba, “Pendeteksian Citra Pengunjung Menggunakan Single Shot Detector untuk Analisis dan Prediksi Seasonality”, JuTISI, vol. 7, no. 1, Apr. 2021.
This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial used, distribution and reproduction in any medium.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.