Identification for Red Automobile Based on Cifar-10 Using Machine Learning Models

Abstract

Algorithms were made and improved across time in different application fields to help them better fit and solve problems that human encounter. Specifically, this paper focuses on the identification mission, that is commonly needed and used by governmental facilities and police departments to track certain objects. However, it is not always that easy, and methods were made to help invent and test these algorithms, along with a vast dataset collected for use, so that they can be examined before putting into real-life use. An existing dataset called Cifar-10 was introduced and chosen, and different methods were introduced and used, to design and examine the accuracy of an identification method. This paper mainly focuses on a red automobile identification model. The experimental results demonstrated the effectiveness of the model. Further usages of similar models will also be applicable with corresponding adjustments, hoping to make it into other similar areas and fulfilling similar goals.