In the past few years, agricultural production studies have gained popularity, showing signs of rapid development. Using different kinds of computer technology, the newest thing to come along makes it easier for farmers to do their work. Many things affect agricultural output, but the efficiency of the seeds is the most important. Seed classification could give us more information about quality work, controlling seed quality, and finding impurities. Automated classification of seeds has been generally done based on factors like colour, texture, and size. Most of the time, specialists do this by looking at the samples, which is very much time-consuming. Adaptation of technologies for automated classification of seeds can be helpful in this regard. Fortunately, there are good number of research already carried out using Deep Neural Network (DNN) around the globe. In this paper, we provide a review of seed classification techniques with a strong focus on DNN. The goal of this research is to create a system for categorizing seeds based on visual and morphological traits.