Proceedings of the 4 rd International Conference of Recent Trends in Environmental Science and Engineering (RTESE'20) Niagara Falls, Canada Virtual Conference – November, 2020 Paper No. 163 DOI: 10.11159/rtese20.163 163-1 Reviewing the Trend in Image Processing Techniques Used in the Agriculture Industry. Smriti Sridhar 1 , Rajiv Gupta 2 , Garrick Louis 3 1 MS Student, Department of System Engineering, University of Virginia, Charlottesville, USA. ss4vh@virginia.edu 2 Senior Professor, Department of Civil Engineering, BITS Pilani, Rajasthan, India. rajiv@pilani.bits-pilani.ac.in 3 Associate Professor, Department of Engineering Systems and Environment, Charlottesville, University of Virginia, USA gel7f@virginia.edu Abstract - Agriculture is the backbone of the global economy. With the increasing pressure and demand on agricultural systems to make the industry smarter, there is a need to focus on the void to be filled. A first step to achieving the sustainable development goals in farming, can be to leverage remote sensing to maximize the efficiency on the farm. In this paper, a comprehensive review is dedicated to the state of the art of image processing techniques used in agricultural applications and later encouraging the use of Artificial Neural Networks for more precise feature extraction. The works that we analyze can be categorized in the following application domains of precision farming: a. Crop/Vegetation management b. Land management c. Soil management. With fields growing larger, better monitoring systems are needed for automated management to reduce expenses. Hence, by applying Artificial Intelligence (AI)-powered solutions, farmers will be able to do more work with less effort and improved quality. This paper reviews the concepts, tools and the potential solutions to the agriculture industry and the need for better image processing techniques in remote sensing. Keywords: Deep Learning, Machine learning, image processing, Remote sensing. 1. Introduction While AI sees a lot of direct applications across sectors, it can also bring a paradigm shift in how we see farming today. From detecting plant diseases to predicting the crops that can deliver the best returns, artificial intelligence can help with one of the biggest challenges in agricultural domain: feeding an additional 2 billion people by 2050. [1] Applications of image processing in agriculture can be broadly classified into two categories: firstly, depending on the image understanding techniques and the second is based on the application of the same in the agricultural domain. This review mainly focuses on the trends of image processing methodologies in various domains of agriculture through these years and how they can be modified for the better. We perceive an image’s content based on objects. Once having perceived objects, we link them together by means of a complicated network made up by experience and knowledge. This very step was hardly implemented in image interpretation software or any other modelling used in the state-of-the-art techniques. The image analysis presented here implies dealing with and handling image semantics. In most cases, information important for the understanding of an image is not represented in single pixels but in meaningful image objects and their mutual relations. Procedures for image object extraction which are able to dissect images into sets of useful image objects are therefore a prerequisite for the successful automation of image interpretation. In this paper, we aim to juxtapose three major areas: AI, agriculture, and image processing to analyze the trends in their applications on the agro-industry over these years. To achieve this, first, a study of image processing techniques used in major spatial and spectral feature extraction is conducted. Lastly, we encourage the use of Artificial Neural Network (ANN) models in other applications. The structure of the presented work is as follows: Section 2 depicts the current trends in image processing-based methodologies used. Section 3 gives an overview of the literature on the applications of AI based image processing techniques (mentioned in the previous section) to find agricultural objects of interest for better precision farming. Finally, in Section 4, the concluding inferences and future expectations in the domain.