New Retail Model Based on Population Density Detection
作者：Lin Chen
来源：Journal of Physics:Conference Series
Abstract.
With the popularization of computer technology, the transformation of business models has gradually become one of the main influencing products of computer technology development. The emergence of the “new retail” model means that the traditional retail model has become less and less suitable for the current commercial market. This paper combines the current cutting-edge methods of deep learning in the field of computer vision to optimize the traditional retail model and propose a new retail model. This model can effectively solve the problem of poor human resources in the traditional retail model. Moreover, the new retail model can apply relevant technologies such as MCNN to count relevant important business indexes such as population density, and can also be used for security inspection and intelligent queuing.
1. Introductions
"New Retail" is to encourage relevant enterprises to combine online, offline and mobile channels, and to promote the comprehensive transformation of price consumption to value consumption by the three parties [1]. This evolution of the retail model involves not only the transformation of the consumer's consumption philosophy, but also the innovative changes in traditional retail behavior. Replacing retail formats and supply chains with "new technologies" such as big data and artificial intelligence. Promote physical retail transformation and upgrading with Internet thinking, and improve circulation efficiency and service level with the support of "new logistics".
Many stores use banquets and POS machines to count customer traffic. No matter which method has certain limitations on statistical accuracy. Different from the traditional method of customer traffic statistics, this paper proposes to use the camera of the merchant to implement the statistics and diversion of passenger traffic based on the deep learning method of crowd density detection to improve operational efficiency. By means of the flow data of the queued cashier area, the convolutional neural network is used to realize the reasonable distribution of human flow, and at the same time eliminate the safety hazard caused by crowding when the customer queues.
2. Background
2.1. The definition of "new retail"
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Ma Yun, Chairman of the Board of Directors of Ali Group, believes that the future will be a “new retail” model combining offline, online and logistics. As shown in Figure 1, only the offline, online, and logistics are truly combined to truly create a "new retail." This article collects other definitions of "new retail", and some scholars have defined the concept of "new retail" from different perspectives. Wang Baoyi proposed in 2017 that “new retail” is the return of the essence of retail. It is a comprehensive retail format that better meets the needs of consumers for shopping, entertainment and social multi-dimensional integration in the era of data-driven and consumption upgrades [2]. This is also the first time the industry has proposed to change the concept of logistics to Omni-channel, which has expanded the consumer's shopping model. At the same time, Zhao Shumei proposed in 2017 that “new retail” means that enterprises apply advanced Internet thinking and technology, improve and innovate traditional retail methods, and sell goods and services to final consumers under the guidance of the latest concepts and thinking. The general term for all activities [3].This definition leads to the concept of “Internet ” and proposes the use of technology to develop a “new retail” model.
2.2. The definition of "Omni-channel"
The core of "Omni-channel" is to break the original model of relying solely on offline channels, and to extend and expand the marketing channels of retail enterprises through online and mobile channels, there by relieving the disadvantages of offline channels in terms of time, space and price. “Omni-channel” has the following main features.
2.3. The combination of artificial intelligence and new retail
The essence of the new retail is to reconstruct the "people-goods-field", in which "people" is an important part, we often think of "consumers", "thousands people have thousands ideas" and "consumer experience" and other factors. But we ignored an important factor named "person", also means human traffic. Whether it is Internet technology or the Internet of Things, Big Data, Artificial Intelligence Technology in the retail industry, the ultimate goal is providing consumers with a better experience. This paper collects and summarizes the methods of population density detection proposed by the top conferences in recent years. At the same time, the crowd density detection algorithm which is most suitable for the new retail model is summarized, and the algorithm is used to realize the function of human flow grooming and improve the efficiency of the cashier area.
3. Architecture.
In this part, a convolutional neural network is used to analyze the crowd count. This paper mainly introduces the algorithm structure and application scenario of the convolution neural network. Retail scenes are often faced with large crowd density irregular passenger flow and so on. Convolutional neural network can be used to process the video image frame captured by the camera in order to control the flow of people.
3.1. Introduction of Convolutional Neural Network.
Convolutional neural network is one of the most commonly used network structures to solve computer vision problems in the field of deep learning of artificial intelligence [4]. The applicati
剩余内容已隐藏，支付完成后下载完整资料
英语译文共 4 页，剩余内容已隐藏，支付完成后下载完整资料
资料编号：[438726]，资料为PDF文档或Word文档，PDF文档可免费转换为Word
以上是毕业论文外文翻译，课题毕业论文、任务书、文献综述、开题报告、程序设计、图纸设计等资料可联系客服协助查找。
