International Journal of Electrical and Computer Engineering (IJECE) Vol. 14, No. 2, April 2024, pp. 1730~1738 ISSN: 2088-8708, DOI: 10.11591/ijece.v14i2.pp1730-1738  1730 Journal homepage: http://ijece.iaescore.com Convolutional neural network for estimation of harvest time of forage sorghum (sorghum bicolor) cultivar samurai-1 Kahfi Heryandi Suradiradja 1 , Imas Sukaesih Sitanggang 1 , Luki Abdullah 2 , Irman Hermadi 1 1 Department of Computer Science, Faculty of Mathematics and Natural Science, IPB University, Bogor, Indonesia 2 Department of Nutrition and Feed Technology, Faculty of Animal Science, IPB University, Bogor, Indonesia Article Info ABSTRACT Article history: Received May 17, 2023 Revised Sep 26, 2023 Accepted Nov 4, 2023 One of the economic alternatives to improve the quality of ruminant feed is combining grass as the main feed with high-protein forages such as sorghum. To get a quality sorghum harvest during the period, it must be right when it has good biomass content, nutrients, and digestibility. The problem is that measuring quality in the laboratory has additional costs and time, which is not short, causing delays. An approach with machine learning using a convolutional neural network can be a better solution. This research uses a convolutional neural network algorithm with the right architecture to estimate sorghum harvest time from imaging results of unmanned aerial vehicles. The stages of this research include data collection, pre-processing, modeling, and finally, the evaluation stage. This research compares the results of several convolutional neural network (CNN) algorithm architectural models: simple CNN, ResNet50 V2, visual geometry group-16 (VGG-16), MobileNet V2, and Inception V3. The result is determining the CNN algorithm architectural model that can estimate sorghum harvest time with maximum accuracy. The best result is the simple CNN architectural model with an accuracy of 0.95. This research shows that the classification model obtained from the CNN algorithm with a simple CNN architecture is the choice model for estimating sorghum harvest time. Keywords: Convolutional neural network Estimated harvest Forage sorghum Machine learning Sorghum bicolor This is an open access article under the CC BY-SA license. Corresponding Author: Kahfi Heryandi Suradiradja Department of Computer Science, Faculty of Mathematics and Natural Sciences, IPB University Jl. Agatis, IPB Campus Darmaga Bogor 16680, Indonesia Email: kahfi.heryandi@apps.ipb.ac.id 1. INTRODUCTION Animals, feed, and land are the main concerns in the livestock industry. Grass is the core feed in the cattle breeding industry, but it has a high crude fiber content and low protein, causing low livestock digestibility. Increasing nutrition's value must be combined with additional feed or concentrated feed such as corn, dregs, and other similar feeds, but of course, this will increase production costs. According to [1], there is a solution to improve livestock performance and reduce ration production costs by lowering concentrate ingredients by providing quality forage by combining main feed and Gramineae, one of which is sorghum. In Indonesia, sorghum plants can be planted in marginal and dry areas [2]. The best solution to provide forage for ruminants is by choosing sorghum as the main feed [3]. According to [4], sorghum has a greater biomass than corn. Forage sorghum can be harvested from the flowering stage [5]. One of the uses of information technology support in agriculture and animal farming is applying one of the algorithm models from machine learning, namely computer programming using historical data [6]. According to [7], machine learning methodology generally involves learning from training data. Machine learning can assist with quick plant phenotyping, agricultural monitoring, and yield prediction [8]. Previous studies using machine learning