Automatic Classification of Aerial Imagery

  • Jharna Majumdar Aeronautical Development Establishment, Bangalore
  • Lekshmi S. Aeronautical Development Establishment, Bangalore
  • B. Vanathy Aeronautical Development Establishment, Bangalore
Keywords: Aerial imagery, automatic image classifier, image information, surveillance, image processingalgorithms, image classification


The aerial imagery obtained from reconnaissance platform is voluminous and the defenceforces rely on image information to perform intelligent tasks. The application of a welldesigned automatic image classifier would enhance the end results of different high levelapplications thereby abridging the effort of a human analyst. Automatic image classifierscould be designed using a training data set for supervised learning or using an unsupervisedlearning. In this paper, a method, which combines both unsupervised and supervised methodsof learning is proposed.

Author Biographies

Jharna Majumdar, Aeronautical Development Establishment, Bangalore
Ms Jharna Majumdar received her BTech (Hons) in Electronics and ElectricalEngineering and Postgraduate Diploma in Computer Technology from Indian Instituteof Technology, Kharagpur, in 1969 and 1970, respectively. She received her PhDin Electrical Engineering in 1980. She has about 34 years of research experience.She worked as Research Scientist in the area of robotics and automation at theInstitute of Real Time Computer Systems and Robotics, Karlsruhe, Germany, during1983 to 1989. Currently she is a Scientist and Head of Aerial Image ExploitationDivision, Aeronautical Development Establishment (ADE), Bangalore. Her researchareas include computer vision, artificial intelligence, and aerial image exploitation.
Lekshmi S., Aeronautical Development Establishment, Bangalore
Ms Lekshmi S. graduated in Mathematics and postgraduated in Computer Sciencein   1995 and 1997, respectively. She received her PhD in Computer Science fromUniversity of Kerala in 2003. She worked as a Research Fellow for ISRO projectand STEC project and was research associate of ADE (DRDO). Presently, she isworking as a scientist in Aerial Image Exploitation Division, Aeronautical DevelopmentEstablishment (DRDO), Bangalore.   Her research areas are: Fractals, patternrecognition, astronomical image analysis and aerial image exploitation.
B. Vanathy, Aeronautical Development Establishment, Bangalore
Ms B.Vanathy received her BE (Electronics and Comm Engg) and ME (ComputerScience) in 1997 and 1999, respectively. Presently, she is working as a Scientistin Aerial Image Exploitation Division, ADE, Bangalore. Her research areas are:Registration, mosaicing, and aerial image exploitation.
How to Cite
Majumdar, J., S., L., & Vanathy, B. (2007). Automatic Classification of Aerial Imagery. Defence Science Journal, 57(6), 773-778.
Aeronautical Systems