Contact Frank Gyan Okyere Okyere
Areas of expertise
- Applied Informatics
- Digital Agriculture
Frank Gyan Okyere has Master's Degree in Biosystems Engineering specializing in Machine Learning application to plant research. His research work was in plant disease detection and rate of spread analysis using machine learning. He utilizes tools such as regression, support vector machines, k-means clustering and other supervised and unsupervised models in detection, identification, classification and quantification of plant phenotypic traits. His current which involves plant disease detection using image segmentation methodology, quantification using k-means clustering and rate of spread (using SVM) earned him best paper presentation in the Greensys 2019 conference. Prior to joining Cranfield University, he schooled at Gyeongsang National University(GNU)- South Korea and has internship experience with Electronic Telecommunication Research Institute(ETRI) South Korea, where he helped established a semi automated Glasshouse at the Department of Bisoystems Engineering research center(GNU). He has four years industrial experience with GRATIS Foundations(Ghana) where he worked as the Production Engineer.
Frank has worked on a number of projects in the past working in the areas of agricultural engineering (mechanical) and plant science(plant phenotyping). He worked on the ETRI - GNU project(2018-2020) where he developed a machine learning model to predict plant disease occurence, its spread rate and the impact of environmental conditions on the spread rate. He also worked on the SMART FARM project(2017-2023) where he studied on improving the performance of the planting device of a vegetable transplanter. He also worked on projects such as Composting of livestock manure, GPS and GIS application to agriculture and Tillage operational analysis based on soil physical and machine mechanical variability. His last project before joining Cranfield University was on Baler Machine where he conducted a research on the influence of operational properties and material physical characteristics on mechanical cutting properties of corn stalks. He designed new cutting system for baler material harvesting. Frank has published seven papers in peer reviewed journals.
His research interest lies in the use of computer vision and computational analysis in plant phenotyping where he is applying image processing techniques and data science theories to analyze plant phenotyping traits.
Frank Gyan Okyere is currently working as a PhD candidate at Cranfield University on the OCP Group project. His project title is 'Image analysis for plant phenotyping–machine learning based methods for analysis of multi-model and multi-dimensional remote sensing data from high-throughput plant phenotyping'. The main objective is to develop an automated protocol system to analyse plant's phenotypic traits and make future predictions of nutrients and water status. He is currently conducting most of his research works at Rothamsted Research and the glasshouse at Cranfield University. After the automated system has been developed, it will be used to validate phenotypic traits of crops grown on the field in University of Mohammed VI Polytechnic (UM6P). Using the Lamtec gantry scanalyzer which has various sensors including RGB, Hyper-spectral, multi-spectral, 3D and laser. The main objective is for development and validation of pattern recognition and machine learning models to predict crops quality traits based on phenotypic data. In this project, crops trial will be developed on the field at Rothamsted and greenhouse at Cranfield where treatments of the trials will be on nutrient and water deficiencies. We are currently co-developing automated sensing and image analysis framework for knowledge interpretation, anomaly detection, clustering and predictive analytics based on dynamically evolving models. Through this project, a web- based feature extraction and image analysis modules will be implemented to perform raw data pre- processing on the platform.