Predicting the performance of the crops to estimate production and determine the amount of agricultural inputs needed is crucial in precision agriculture. In the present study, a fast, accurate, and inexpensive method for estimating sugar beet yield with and without leaves was presented. Images were captured from the crop in both conditions and then, six morphological features including area, perimeter, major axis length, minor axis length, equivalent diameter and centroid were extracted from the images. Pearson correlation coefficient analysis was used to select the best effective feature. Features with a correlation coefficient greater than 0.7 were considered as effective features. Accordingly, for the two conditions of with and without leaves, area and perimeter characteristics were selected, respectively. To check the accuracy of the linear weight estimation equations, equations were given to the online crop recognition system (in two modes) and the root weight of the sugar beet was estimated immediately by tractor movement. Results showed that there was a correlation coefficient of 0.84 and 0.95 between the actual- and estimated weight in both with and without leaves conditions, respectively. According to the results, image processing system can be used as an estimation system for online sugar beet yield prediction.
- حق عضویت دریافتی صرف حمایت از نشریات عضو و نگهداری، تکمیل و توسعه مگیران میشود.
- پرداخت حق اشتراک و دانلود مقالات اجازه بازنشر آن در سایر رسانههای چاپی و دیجیتال را به کاربر نمیدهد.