Indonesian Journal of Electrical Engineering and Computer Science Vol. 33, No. 2, February 2024, pp. 990~998 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v33.i2.pp990-998  990 Journal homepage: http://ijeecs.iaescore.com Maize seed variety identification model using image processing and deep learning Seffi Gebeyehu 1 , Zelalem Sintayehu Shibeshi 2 1 Department of Software Engineering, Faculty of Computing, Institute of Technology, Bahirdar University, Bahirdar, Ethiopia 2 Department of Computer Science, Rhodes University, Makhanda, South Africa Article Info ABSTRACT Article history: Received Aug 9, 2022 Revised Oct 31, 2023 Accepted Nov 6, 2023 Maize is Ethiopia’s dominant cereal crop regarding area coverage and production level. There are different varieties of maize in Ethiopia. Maize varieties are classified based on morphological features such as shape and size. Due to the nature of maize seed and its rotation variant, studies are still needed to identify Ethiopian maize seed varieties. With expert eyes, identification of maize seed varieties is difficult due to their similar morphological features and visual similarities. We proposed a hybrid feature-based maize variety identification model to solve this problem. For training and testing the model, images of each maize variety were collected from the adet agriculture and research center (AARC), Ethiopia. A multi-class support vector machine (MCSVM) classifier was employed on a hybrid of handcrafted (i.e., gabor and histogram of oriented gradients) and convolutional neural network (CNN)-based feature selection techniques and achieved an overall classification accuracy of 99%. Keywords: CNN Deep learning HOG Maize seed Variety identification This is an open access article under the CC BY-SA license. Corresponding Author: Zelalem Sintayehu Shibeshi Department of Computer Science, Rhodes University Makhanda, South Africa Email: z.shibeshi@ru.ac.za 1. INTRODUCTION Properly utilizing various seed types is one of the most significant aspects of any agricultural development system. This technique has received emphasis in recent times as it is expected to harness food security to the increasing population in the world, for achieving better harvest execution and production yield, the identification of high production yielding seed varieties deprived of vital characteristics such as flooding, breaking obstruction, insects, and diseases. The production of a high-quality seed variety is the basis of any effective agricultural system since agriculture is the economic backbone of most countries worldwide. Basic requirements such as food, clothing, shelter, and medicines increase as the population grows. Advancement in agriculture is needed to meet these needs, especially for countries with large populations [1]. Regarding production output, maize, also called corn, has become the major grain crop worldwide. As mentioned in [2], between 2017 and 2019, the total production of maize worldwide was 1,137 metric tons, which is 39% of all cereals produced in those years. Maize is considered the cheapest calorie source compared to the other grains and now constitutes Ethiopia ’s highest calorie consumption and production share [2]. Due to the significant variation in price and resemblance in form, the impurity of maize seed varieties creates many challenges. Identification of the type of maize seeds is an essential aspect of the development of the maize industry. In Ethiopia, maize has traditionally been a significant food source for many people. The majority of Ethiopian farmers are known for cultivating maize crops. As a result, maize has become the second dominant crop next to teff (Eragrististeff) among the cereals grown in Ethiopia [3].