Desarrollo de una aplicación web mediante acceso facial con deep learning para el control de mercadería de la ferretería Jeroba

Abstract

All kinds of software, whether system, application or programming, together with hardware that is the part of a computer system to which we have access manually, have helped us improve digital communication, in real time, documentation, active programming, text file converted, file information, and much more. A series of activities or habits that in some way improve our quality of life and will surely continue to do so. Software development is a very common activity these days and the big companies working on it like Microsoft, IBM and Oracle are constantly working and innovating as they have to update their products and use the latest technology to meet the needs. of the growing demand, expect consumers to have new features, which means that these brands have to put a lot of effort into research and development, the ability to deeply understand what users want, but also the excess of creativity to impress them and create an even more impressive product. This often leads us to believe that there will be nothing new to do in software in the future, whatever the field or purpose of product development. Software is undoubtedly a very important tool in the world, it is part of everyone's life and their devices and it will surely last a long time, bringing more and more useful and surprising new features that will further simplify our personal and professional lives. For now, we just need to know that technological developments will affect us both positively and negatively on a day-to-day basis, and it all depends on how we use and allocate the tools that we have at hand every day. Deep learning has drawn a lot of attention due to its ability to be used in different types of real-world applications, it can be successfully applied to large amounts of data for insight discovery and application. An artificial neural network consists of an input layer, at least one intermediate layer, and an output layer for information processing. The data arrives at the input layer as an input vector, is then weighted by the artificial neurons in the middle layer, and finally forms a specific pattern in the output layer. The more layers an artificial neural network has, the more complex tasks it can handle. The higher performance of deep learning compared to previous AI techniques is proof of the immense media attention it has received, as well as the research interest in the topic. However, for this technology to reach its full potential, there are some weaknesses that it shows first. To achieve the main objectives of the proposal, a detailed analysis of the functional and non-functional requirements was carried out, followed by the identification of manual processes by the personnel of the operational area. Based on the analysis of requirements, several modules emerged such as customer, product, sales, among others. For the development of the proposal, the SCRUM methodology was selected, which complies with phases that facilitate its realization. In addition, the php and Python programming languages were used, as well as MySQL as a database administrator.