Performance evaluation of Googlenet, Squeezenet, and Resnet50 in the classification of Herbal images

  IJETT-book-cover  International Journal of Engineering Trends and Technology (IJETT)          
  
© 2021 by IJETT Journal
Volume-69 Issue-3
Year of Publication : 2021
Authors : M.A. Muthiah, E. Logashamugam, N.M. Nandhitha
DOI :  10.14445/22315381/IJETT-V69I3P234

Citation 

MLA Style: M.A. Muthiah, E. Logashamugam, N.M. Nandhitha "Performance evaluation of Googlenet, Squeezenet, and Resnet50 in the classification of Herbal images" International Journal of Engineering Trends and Technology 69.3(2021):229-232. 

APA Style:M.A. Muthiah, E. Logashamugam, N.M. Nandhitha. Performance evaluation of Googlenet, Squeezenet, and Resnet50 in the classification of Herbal images  International Journal of Engineering Trends and Technology, 69(3),229-232.

Abstract
Computer-aided classification of medicinal herbs is of major concern in medicinal research. The real challenge lies in the complexity and variability of plants belonging to the same species. Conventional approaches involve feature extraction and classification. Feature extraction is not that effective due to the shift variance of plots. Hence an exhaustive research technique is needed to perform the classification of medical herbs. Convolutional Neural Networks are the recently accepted paradigm for classification with the help of pre-trained neural networks. In this paper feasibility of GoogleNet, SqueezeNet, Resnet 50 for the classification of medicinal herbs is studied. Its is found that Resnet 50 provides high sensitivity for both trained and test dataset.

Reference
[1] Nur Azida Muhammad, Amelina Ab Nasir, Zaidah Ibrahim, Nurbaity Sabri., Evaluation of CNN, Alexnet, and GoogleNet for Fruit Recognition, Indonesian Journal of Electrical Engineering and Computer Science, 12(2) (2018) 468~475
[2] K.K. Sudha, P. Sujatha., A Qualitative Analysis of Googlenet and Alexnet for Fabric Defect Detection, International Journal of Recent Technology and Engineering (IJRTE), ISSN: 2277-3878, 8(1) (2019) 86-92.
[3] Gurnani Ayesha, Viraj Mavani, Vandit Gajjar, Yash Khandhediya., Flower Categorization using Deep Convolutional Neural Networks, (2017) 4321-4324.
[4] Inkyu Sa, ZongYuan Ge, Feras Dayoub, Ben Upcroft, Tristan Perez, Chris McCool., Deep Fruits: A Fruit Detection System Using Deep Neural Networks, Sensors, (2016).
[5] N. Selvarasu., Alamelu Nachiappan., and N.M. Nandhitha., Abnormality Detection from Medical Thermographs in Human Using Euclidean Distance Based Color Image Segmentation.,International Conference on Signal Acquisition and Processing., IEEE, (2010).
[6] A. Bindhu, Dr. K. K. Thanammal., Analytical Study on Digital Image Processing Applications, IJETT International Journal of Computer Science and Engineering, 7(6) (2020) 4-7.

Keywords
GoogleNet, SqueezeNet, Resnet 50, Herbs, Classification, Sensitivity, Specificity