Scientists engaged in the field of Artificial Intelligence (AI) say it is one of the branches of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence. It is an interdisciplinary science with multiple approaches. But advancements in machine learning and deep learning are creating a paradigm shift in almost every sector of the technology-leaning industry.
Moreover, AI is also capable to perform tasks that ordinarily require human intelligence. Many of these artificial intelligence systems are powered by machine learning, some of them are powered by deep learning while others are powered by very boring things like rules, they added.
Recently, the first of its kind AI Workshop and Expo in Ethiopia was held. Assistant Professor of Information Technology Institute of Jimma University, Getachew Mamo (PhD) said that currently, technology has brought new advancements that can successfully and accurately identify disease types and suggest most relevant remedies for specific plant and animals. One of such advancements is artificial intelligence. It can help to find disease and related genes.
Meaningfully, AI is an entity that enables to receive inputs from the environment, interpret & learn from such inputs and exhibit related and flexible behaviors and actions that help the entity achieve a particular goal or objective over a period of time. There are various processes through which a machine uses inputs. The processes that a machine uses to learn are known as algorithms, he said.
Currently, deep learning algorithms are widely used in the area of complex, features since it uses artificial neural network architectures that contain a large number of processing layers that enables it to handle complexities of the problems. Deep learning approach studies features of the problems from a give inputs with minimum human intervention. Thus, machine learning algorithms have been practically applied in various sectors that are complex by their features including the agricultural sector.
Specially, in crop production interventions, traditionally the most widely used approach for disease and pest control is spraying pesticides in the farm uniformly without clear identification of the problem. Accordingly, it has environmental impacts such as ground water contamination, residue on crop products, impacts on wild-life and eco-system among others, he said.
However, AI-based applications can manage these agro-chemical inputs in terms of place and time where pesticides will be applied on affected areas instead of uniform applications. But AI Intervention in Agricultural Sector; accuracy of identification plays a great role to save consequence costs and damages as a result of wrong decision. Thus, smart agriculture is important for tackling the challenges of agricultural production in terms of productivity, environmental impact, food security and sustainability.
Therefore, the review is categorized based on topic of the research, the major topics being application of AI in disease detection, pest control, weed detection, yield prediction and nutrient deficiency detection.
One of the most serious concerns in agriculture is disease and pest control. But, AI can help alleviate this by targeting agrochemical inputs in terms of place and time where pesticides will be applied to affected areas instead of uniform application.
The main threat to crop production is the negative effects of weeds on crops which include competition for water, light, nutrients, increased production costs, difficulty in harvesting, and depreciation of product quality, increased risk of pests and diseases and decrease in the commercial value of cultivated areas. AI can lead to accurate detection of weeds with low cost using sensors. Such weed detection systems will enable the development of tools and robots that can destroy weeds, he said. This can significantly minimize the need for herbicide downsizing environmental side effects, he said.
In addition, crop yield prediction is an essential task for decision-makers at national and regional levels for rapid decision making. An accurate yield prediction model can help farmers to decide as to what and when to grow. It uses image recognition and Deep learning (DL) algorithms to detect about 400 damage occurrences in 30 crop types. Besides, it also allows weather prognosis.
Nonetheless, it is obvious that agriculture is the practice of farming, including cultivation of the soil for the growing of crops and the rearing of animals to provide food, wool, and other products to sustain the global human population.
According to Dr. Getachew, the world’s population is expected to grow to almost 10 billion by 2050.Hence, the demand is agricultural production is expected to be boosted. In most African countries, 60 percent or more employees are engaged in agriculture. Particularly, in South of the Sahara – about 30 percent of GDP is from agriculture and a significant proportion of export is based on agriculture, he said .
The same is true in Ethiopia. Agriculture contributes 37 percent of GDP. Seventy three percent of the total employees are engaged in the sector. It supports 70 percent of foreign exchange too. Moreover, the country has diverse agro-ecology- suitable for various farming systems so that it allows a diverse biological wealth of crop and livestock production.
However, agricultural practices are predominantly characterized by a high level of subsistent production and lack of production for market. Farming relies on traditional methods. Oxen-driven tilling is still there. Due to these and other reasons food insecurity is one of the main challenges of the country. Thus intensifying AI technology researches will ease challenges confronting the agricultural and animal production, he said.
Diseases, Pests and drought, soil degradation and acidity, poor farming skills are challenges of the agricultural sector too.
Obviously, crop production is widely practiced in Ethiopia. It shares 14 percent of GDP. Staple crops dominated crop production in Ethiopia include: cereals, pulses, oil seeds, roots and tubers, vegetables and coffee. Cereals, pulses, oil crops constituted 78 percent, 15 percent, 7 percent of the cultivated area respectively.
Their total grain production is 85 percent, 12 percent and 3 percent of the country respectively. Smallholders account for 96 percent of the total area cultivated in Ethiopia. The five cereal crops are Teff, wheat, maize, sorghum and barley .However, their productivity are very low.
Importantly, the main challenges for global production are plant Diseases and Pests.
Wheat Rusts are among the most dangerous fungal diseases of wheat worldwide. The three most economically dangerous and common rust diseases globally are Leaf or brown rust caused by Puccinia triticina Eriks, Stem or black rust caused by Puccinia graminis f. tritici Erik and Stripe or yellow rust caused by Puccinia striiformis f. Tritici Eriks.In Ethiopia stem and yellow rust diseases are the major biotic constraints articular crops.
Barley is among the primary cereal crops grown in Ethiopian highlands .The most widely distributed and economically dangerous diseases are sald Net blotch, Spot blotch, Leaf rust ,Smuts and Eyespot.
Teff is commonly cultivated throughout Ethiopia as a staple cereal crop. The most commonly known Teff diseases are: Teff rust , head smudge and damping-off . Moreover, maize is one of the most important staple food crops in sub-Saharan Africa (SSA) .Currently, the number of diseases has increased and reached up to 65 in number.
The major diseases identified/or recognized are Gray leaf spot ,Turcicum leaf blight, Common leaf rust Schw and Maize streak virus.
Coffee contributes 12 percent of the agricultural output, 10 percent of the government revenues, more than 60 percent of the foreign exchange earnings. The four main coffee diseases; Coffee Leaf Rust (CLR) coffee wilt disease (CWD) Coffee Berry Disease (CBD) and Coffee Berry Borer (CBB)
Furthermore, Ethiopia is the highest livestock population in Africa. The sector has contributed up to 47percent of agricultural GDP. It contributed for nearly 20 percent of total GDP, and 20 percent of national foreign exchange earnings in 2017. But productivity of livestock is very low in Ethiopia due to animal diseases and lack of feeds.
The three most important diseases of cattle are foot and mouth disease (FMD), contagious bovine pleuropneumonia (CBPP), and lumpy skin disease based on their impact on rural households .The productivity of both crop and animal has being negatively affected by various diseases and pests, he said
BY ALAZAR SHIFERAW