Browsing by Author "Fue, Kadeghe"

Now showing 1 - 3 of 3

Analyzing usage of crowdsourcing platform Ushaurikilimo' by pastoral and agro-pastoral communities in Tanzania
(International Journal of Instructional Technology and Distance Learning, 2016-12) Fue, Kadeghe; Geoffrey, Anna; Mlozi, M.R.S.; Tumbo, Siza, D.; Haug, Ruth; Sanga, Camilius, A.
Earlier studies report that agricultural extension service for livestock keepers in Tanzania is not effectively offered. ‘Ushaurikilimo’ which is a crowdsourcing platform consisting of a Web and Mobile based agro-advisory system. It is a system which complements the traditional agricultural extension provided to pastoralists and agro-pastoralists. Mobile crowdsourcing for agricultural extension service is an emerging approach to address some of the problems experiencing in traditional agricultural extension service. This study explored the information seeking pattern of livestock keepers who are using ‘Ushaurikilimo’. In total 1739 questions are in ‘Ushaurikilimo’. Out of 1739 questions and answers, the study concentrated on 1312 questions since 427 questions related to forestry. Out of 1312 questions submitted to ‘Ushaurikilimo’ via livestock keepers’ mobile phones, 605 (47%) questions relate to livestock and 53% relate to crops. Most livestock keepers asked questions with keywords related to chicken, pigs and milk from ‘Ushaurikilimo’ knowledge base. Further data mining analysis showed that the following keywords are the most queried information by livestock keepers: poultry management, poultry equipment and accessories, hatching equipment, feed equipment, feed storage, feed manufacturing machinery, product handling/transport equipment, quality testing equipment and energy saving equipment, milk processing, housing and environment, building materials and equipment, feeds and feeding, food preservation, feed additives and dairy products. In this study, the pattern of information seeking behavior of livestock keepers matched the pattern which has been reported earlier by other researchers who explored the information seeking behavior of livestock keepers who are using other sources of information such as newspapers, television, radio, farmers’ friends and extension agents. One peculiar result from this study is that the average response time after the question had been assigned to an expert to answer was 32.49 hours. Thus, the crowdsourcing platform, web and mobile based agro-advisory system proved to be effective compared to conventional agricultural extension methods. This calls for a need to scale up ‘Ushaurikilimo’ to complement the traditional agricultural extension service in Tanzania.
Visual control of cotton-picking Rover and manipulator using a ROS-independent finite state machine
(2019 ASABE Annual International Meeting, 2019-07) Fue, Kadeghe; Barnes, Edward; Porter, Wesley; Rains, Glen
Small rovers are being developed to pick cotton as bolls open. The concept is to have several of these rovers move between rows of cotton, and when bolls are detected, use a manipulator to pick the bolls. To accomplish this goal, each cotton-picking robot needs to accomplish three movements; rover must move forward/backward, left/right and the manipulator must be able to move to harvest the detected cotton bolls. Control of these actions can have several states and transitions. Transitions from one state to another can be complex but using ROS-independent finite state machine (SMACH), adaptive and optimal control can be achieved. SMACH provides task level capability to deploy multiple tasks to the rover and manipulator. In this research, a cotton-picking robot using a stereo camera to locate end-effector and cotton bolls is developed. The robot harvests the bolls using a 2D manipulator that moves linearly horizontally and vertically. The boll 3-D position is determined by calculating stereo camera parameters, and the decision of the finite state machine guides the manipulator and the rover to the destination. PID control is deployed to control rover movement to the boll. We demonstrate preliminary results in a direct-sun simulated environment. The system achieved a picking performance of 17.3 seconds per boll. Also, it covered the task by navigating at a speed of 0.87 cm per second collecting 0.06 bolls per second. In each mission, the system was able to detect all the bolls but one.
The YEESI Lab Dataset (1.0)
(Zenodo., 2022) Fue, Kadeghe; Barakabitze, Alcardo; Geofrey, Anna; Lebalwa, Bertha; Lyimo, Neema; Mwaipaja, Faraja; Jonathan, Joan; Mbacho, Susan; Sanga, Camilius; Rains, Glen
The Tanzanian agriculture industry faces a great challenge caused by pests and diseases threatening food security. Pests such as tomato leaf miners, aphids, fall armyworms (FAW), and bean leaf miners devastate crops. Also, diseases such as maize streak virus, early blight, Powdery mildew, Leaf spot, Rusty brown leaf, foliar disease, Bacterial Wilt, Blossom end rot, Flower abortion, Leaf Curl and Black rot have caused the crop failure that leads to yield reduction. So, precisely and accurately detecting such pests and diseases to improve agriculture productivity in the country is paramount. However, manual detection is cumbersome, time-consuming and costly. So, automating the procedure using machine vision technologies is necessary for sustainable prosperous agriculture. Therefore, this dataset presents the first Tanzanian agricultural classification dataset that contains 7992 healthy and unhealthy crops images (maize, beans, green peppers, onions, okra, watermelons, sunflowers, African eggplants, tomatoes, Chinese cabbage, hot peppers, wheat, leaf kale and cabbage). Images were collected in real-world conditions in Morogoro, Tanzania, in August and September 2022, using smartphones and professional GoPro Hero 9 cameras. The dataset is called YEESI Dataset. It is used as Open Data. The authors expect this dataset to revolutionize applications of Artificial Intelligence (AI) in agriculture for evaluating classification models related to crop pests, diseases and weed problems from open data.