Farm monitoring and control using drones with object detection capabilities.
Lead Applicant Organization Name
African University of Science and Technology, Abuja
Lead Applicant Organization Type
If part of a multi-stakeholder entity (i.e. team), provide the names of other organizations and types of stakeholders collaborating with you.
Rise Network, Lagos
Website of Legally Registered Entity
How long have you / your team been working on this Vision?
Lead Applicant: In what city or town are you located?
Lead Applicant: In what country are you located?
Your Selected Place: what’s the name of the Place you’re developing a Vision for?
What country is your selected Place located in?
Describe your relationship to the place you’ve selected.
Many are farmers, as nearly 80 percent of Nigeria's northern population works in the agriculture industry. With an increase in Boko Haram attacks and the displacement of nearly two million Nigerians, agricultural production has plummeted and staple food prices have sky-rocketed.
Northeast Nigeria now faces one of the world's worst food security crises, with around 3.8 million people who will face critical food insecurity and around 7.7 million in need of life-saving humanitarian assistance in 2018. The Northern has become a source of income directly from my family.While I was 8, my mother began selling yam tubers as a means of livelihood for our family and up till now. It was our family business and everyone in the family was deeply involved in the yam trading business. My mother travels regularly to farms to meet farmers to purchase yam tubers for merchandise. The cost of yam increased significantly and has continued to increase. This directly affected my family and likewise society. We discovered the increase in the price of yams was due to the great loss of crops in the farms. I decided to help the farmers to control this loss. With the loss in the farm combated my family and society would have plenty to eat and earn. The solution arises from a passion to solve a personal and family issue.
The idea of using drones and machine learning arose during our master's degree program in computer science. Haven carried out my research in drone technology and machine learning and considering my history from an agricultural-dependent family and problems we have farmers complain about we decided to solve these challenges of loss of crops and breakout of diseases.
Describe the People and Place: Provide information that would be helpful for an outsider who has never been there and may have no context about this Place to better understand the area.
Northern Nigeria is primarily occupied by the Hausa. Hausas are the largest ethnic group in all of West Africa, and of which 30 percent of all the Hausas can be found in the north, particularly the northwest region of Nigeria, an area known as "Hausaland”.It is worrisome that the Hausa people in northern part of Nigeria seem to be lagging in terms of socio-economic development, as the region is considered with the highest rate of poverty and unemployment, least educated as well as more illiterates, more homeless kids and street urchins (almajirai), more gender disparity with females being subdued in almost every sphere of the society, higher vulnerability to disease, and many of these negative impacts. The Hausa community and its culture are said to be complex in nature possessing different aspects of life that are on the extremes: whether in urban and rural settings, agriculture and highly specialized craft production etc. Agricultural sustainability in northern Nigeria requires flexibility in both ecological management as well as economic activity. Rainfall occurs only seasonally – and there is a pronounced dry season – however, rainfall is often intensive when it does come, making it necessary for farmers to employ soil moisture conservation techniques. The main crops grown in the region are millet, sorghum, and cowpea, while groundnut and sesame are significant minor crops. Wild foods also serve as an important supplement to the diet, especially during times of food shortage. 84% of people living in rural areas cook their food over an open fire, heated by wood or straw. The bulk of crops are grown during the rainy season which begins in June or July when temperatures are warmer. There has traditionally been a division between sedentary farmers made up of the Manga and Hausa people, and the nomadic pastoralists known as Fulani, however this has diminished in recent times. Historically, development plans for this region have focused on the use of imported technology and irrigation schemes, while neglecting traditional farming practices of the region.These traditional practices generally focus on the close integration between the raising of livestock and farming,and have been studied in detail in the Kano Close Settled Zone of Northern Nigeria. The soils in the northern region of Nigeria are categorized as reddish brown or brown soils of the semi-arid and arid regions. They are also known as tropical ferruginous soils and are considered to be comparable to Ferric Luvisols. These are sandy soils that are made up of about 85% sand. Their pH values range between 6.0 and 7.0, and their bulk densities are about 1.4 g/cm3.
What is the approximate size of your Place, in square kilometers? (New question, not required)
What is the estimated population (current 2020) in your Place?
Challenges: Describe the current (2020) and the future (2050) challenges that your food system faces.
Nigeria is facing major challenges with high population growth, a high number of people living in extreme poverty, rapid urbanization, and stagnating agricultural productivity. Tensions between pastoralists and sedentary farmers affect much of Nigeria’s northern region. Gender inequalities are a major barrier to rural development. The central place that oil exports hold in the Nigerian economy limits diversification towards agricultural exports, which have the potential to contribute to more inclusive and sustainable economic development. Public investments in the agricultural sector are low, resulting in underdeveloped (rural) infrastructure and agricultural services. Underlying institutional drivers include weak institutions, weak links between science and practice, low quality of education, corruption, and non-transparent markets with high transaction costs and high investment risks despite the high (urban) demand for food. As Nigeria is a food-deficit country, the urban demand for cheap food is met through food imports, but there is a mutually reinforcing mismatch between supply and demand at many levels and in many dimensions. Negative feedback loops – between a weak enabling environment, lack of incentives and finance for investment and low agricultural productivity – keep the agri-food sector locked into underperformance.
"Nigeria is said to have 20,000 to 30,000 functional tractor units. But the nation still requires about 1.5 million tractor units to fully boost its food production. This projection is against the backdrop of FAO recommendation of tractorization intensity of 1.5 hp (1.125 kW) per hectare. FAO study further showed that only 1% of farm power is supplied by mechanical means in sub-Saharan Africa, 10% of animal draught power while the remaining 89% is from human labor, "AgricultureNigeria.com. Machinery is important in order to make a larger farm. The people that can afford it should look into investing in one or more tractors. Tractors can be used for trailers, transportation, plowing, tilling, harvesting, and much more. The main task to be mechanized is soil tillage. Tilling the soil takes the most time and is difficult to do by hand.
Address the Challenges: Describe how your Vision will address the challenges described in the previous question.
Dronagro would be used by farms to monitor and control farm assets, track and detect pests and diseases on the farm. Early in the day, a yam farmer in Otukpa would usually want to know the situation of his crops, whether they need extra fertilizer or not, whether some yam mosaic virus has infected any of the yams, or weed gradually colonizing some portion of the nutrient needed by the yams to survive. Getting this data and generating high-quality information from the data is typically difficult for the farmer. With Dronagro, this data can be collected in a few minutes, and high-quality information, visualizations, trends, and patterns can be generated in a couple of minutes. The farmer learns about the disease affecting the yams, predictions are made on the possibilities of new disease infections, the estimated number of weeds on the farms and the estimated number of yams on the farms. This information can be used for disease and pest control.
Dronagro enhances the productivity of farms by sourcing and analyzing accurate farm data, proffering assistance to the farmer by using simple and interactive visualizations. Dronagro in its fervid entity rests on deep learning algorithms, computer vision tasks, affordable drone technology, raspberry pi, and extreme human-computer interaction system. With Dronagro, we determined a way in which drones can collect data from the farm using instance segmentation and object detection algorithms in multiple images captured by the drone on the flight. We discovered and designed effective and exceptional methods of extracting, transforming and loading the data into a seamless algorithm and framework that analyzes and visualizes the data. The visualization provides users a means to interact and discover trends intuitively from the charts and hotlines. This interactive visualization can be accessed using a mobile phone and also a PC.
High Level Vision: With these challenges addressed, now provide a high level description of how the Place and the lives of its People will be different than they are now.
One thing that held us spellbound and makes energy surge in our brains is exponential upshot investing quality endeavor in the provision of adequate solutions, technologies and development could bring about. We are ardent on delivering a technology that will provide food for our people. We want in the current year to record over a 35% increase in productivity and yield in over 200 farms that will use our technology. We want to see an increase in food supplied to refugee camps in Borno, and the cost of food fall by at least 20 percent in the next 2 years. We want to democratize access to drone technology and machine learning for farming in Nigeria and Sub-Saharan African within the next five years. We will be providing high-quality agricultural consultancy service and data analytics that will transform the global agricultural terrain using quality data that we would amass by five years. We would be able to create better disease and pest detection and prediction system that would improve productivity the more.
Full Vision: How do you describe your Vision for a regenerative and nourishing food future for your Place and People for 2050?
Billions of dollars are lost annually to the loss of crops and livestock due to pests, diseases, predators and poor handling. Poor output and productivity lead to increased hunger, poverty, and malnutrition in Nigeria and sub-Saharan Africa. Moreover, statistics show that major agricultural production occurs in the rural and suburban areas in this region. With the advance in drone technology and computer vision, this work implements the state-of-art, Mask R-CNN in the instance segmentation and object detection of essential assets and objects on the farm. The assets include livestock, vegetation, structures, water sources, pond, equipment, farm machines. It detects and classifies weed, plant diseases, pests, and predators. In order to generate a high-quality segmentation mask and object instances in an image, Mask R-CNN was developed. Mask R-CNN adds another branch to the Faster R-CNN. In addition to the bounding box recognition system, a branch for predicting an object mask in parallel was added. It affixes only a bijou overhead to Faster R-CNN, running at 5 fps. Images and videos of the divisions of the farm areas are collected using drone technologies, and in real-time these images and videos are analyzed using a trained model of Mask R-CNN. Cardinal information, such as the presence of pests and diseases, count of livestock and vegetation, etc., are generated. These help the farmer make better and quick decisions. The system provides an interactive data visualization and analytics application that is accessible through mobile devices and PC. A good demo of the app can be found on dronagro.herokuapp.com*. After a systematic evaluation, it was discovered that the system efficiently improves farm yields and outputs favorably. We train personnel for the management and use of the technology on the farm and provide monthly product maintenance. Though the product uses deep technologies, we have designed it to be very simple and flexible. Unskilled and uneducated farmers can easily learn to use and operate the product after a few hours of training. Dronagro makes agriculture simply profound and hugely productive.
Our concept impacts every common Nigerian that depends on food produced in Nigeria to live daily. Food production is favorably increased by making proper and qualitative decisions using accurate data that would ensure the detection and prevention of outbreak of diseases and pests on the farm. This solution received a National Artificial Intelligence Ideation Award in 2019 and has a demo of the analytics application and also was among the 60 projects accepted for Harvard Center for Research and Computation in Society (CRCS) Workshop on Artificial Intelligence (AI) for Social Impact 2020.
How did you hear about the Food System Vision Prize?