A New Approach to Determine the Milk Lines of Silage Corn Using Image Processing Techniques

Corn silage is the main roughage used in dairy cattle rations, and it is used intensively in the feeding of animals such as fattening calves, dry cows, and milking cows, especially in modern livestock enterprises. Therefore, the factors affecting the dry matter (DM) yield (t/ha) and feed value of corn silage play an important role in the profitability of the enterprises. Performing this separation using the milk line in corn may cause an incorrect analysis. In this study, corn, one of the plants grown for silage production, was determined on embedded system platforms (Jetson Nano and Jetson TX2) by using image processing techniques to reveal the milk line of corn and to determine the most appropriate harvest time for corn. First, the images of the corn plant are subjected to preprocessing and segmentation processes. After these processes, the corn-milk line is determined, and the starch and milk parts are distinguished with the help of histogram graphics of both parts. Then, by proportioning the histogram values of the two pieces obtained, the starch and milk ratio of the corn is determined. The ratio of these two values determines the harvest time of the corn. In addition, the study is carried out on two different embedded system platforms and compared. In this study, the determination of the milk lines of the corn plant using image processing methods is important in terms of presenting it in the literature.