Performance Evaluation of Conventional CNN Architectures and Modified ALEXNET for the Classification of Potatoes by Ther
- PDF / 702,506 Bytes
- 9 Pages / 612 x 792 pts (letter) Page_size
- 15 Downloads / 168 Views
HERMAL METHODS
Performance Evaluation of Conventional CNN Architectures and Modified ALEXNET for the Classification of Potatoes by Thermal Imaging M. A. Muthiaha, *, E. Logashanmugama, **, N. M. Nandhithaa, ***, Ch. Kranthi kumara, ****, and Dama Haritejaa, ***** a
Department of Electronics and Communication Engineering, Sathyabama Institute of Science and Technology, Jeppiaar Nagar, 600119 Chennai *e-mail: [email protected] **e-mail: [email protected] ***e-mail: [email protected] ****e-mail: [email protected] *****e-mail: [email protected] Received April 9, 2020; revised May 30, 2020; accepted July 24, 2020
Abstract—Quality assessment of potatoes necessitates a computer aided interpretation technique that uses a nonhazardous and noncontact, Non-Destructive Testing technique for image acquisition and a feature extraction technique for extracting and representing the features and a classifier for performing classification/grading of potatoes. In this paper, InfraRed Thermography is used for image acquisition and Convolutional Neural Networks (CNNs) are used for the classification of potatoes. In this work, thermograms of Normal, Fungus affected potatoes, potatoes with Holes and worst affected potatoes are used. Feasibility of VGGNET, RESNET and modified ALEXNET for classification of potatoes is studied. Performance is measured in terms of sensitivity and accuracy. It is found that RESNET18 outperformed all the other networks in terms of accuracy and it is the only network that classified all the affected potatoes as affected potatoes, i.e. the affected potatoes are not wrongly identified as normal. Keywords: Thermogram, potatoes, VGGNET, RESNET, ALEXNET, Quality assessment DOI: 10.1134/S1061830920090077
INTRODUCTION Potatoes play a vital role in Indian economy as it is the only vegetable exported to foreign countries in large quantities. Before being exported, potatoes are subjected to stringent quality assessment procedure. Computer aided assessment is already in place where radiographs of potatoes are acquired and features are extracted and fed to the neural networks (Zhou, 2016). In CSIR, Chennai, hardware is developed and implemented for radiography based quality assessment of potatoes. However, X-rays cannot penetrate through dense tissues and hence quality assessment is limited by this property. Also X-ray set up is costly and may result in health issues. Hence, it necessitates a Non-Destructive Testing (NDT) technique that is noncontact, nonhazardous and noninvasive in nature (Sangeetha 2017). InfraRed Thermography (IRT) is one such technique that captures the heat emitted from object and maps into thermogram (Maldague, 2000). Here, thermogram is two dimensional function g(x,y) where x,y are the spatial co-ordinates and g(x,y) refers to radiance. The choice of IRT is justified by the fact that all objects above zero Kelvin emit heat. Quantity of heat emitted is dependent on the nature and type of the object. In potatoes, temperature of normal tissue is different from that of a
Data Loading...
