Performance Analysis of Deep Learning Architectures in Classifying Calamansi (Citrofortunella microcarpa)

被引:2
作者
Custodio, Epie F. [1 ]
Hernandez, Alexander A. [1 ]
机构
[1] Technol Inst Philippines, Grad Programs, Manila, Philippines
来源
2022 IEEE 12TH ANNUAL COMPUTING AND COMMUNICATION WORKSHOP AND CONFERENCE (CCWC) | 2022年
关键词
convolutional neural network; deep learning; fruit grading; transfer learning; data augmentation; citrofortunella microcarpa; optimizers; CLASSIFICATION;
D O I
10.1109/CCWC54503.2022.9720788
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
The Philippine lime (Citrofortunella microcarpa) locally known as Calamansi belongs to the citrus family and is among the top four produce of the Philippines exported to different countries. When done manually, postharvest grading of Calamansi is a laborious operation. Thus, an effective method of assessing them according to size, color, and maturity is required to assist calamansi farmers in grading and classifying it non-destructively with precision and accuracy. Deep learning architectures have been useful in a wide range of agricultural applications including classifying, grading fruits, and quality recognition. The goal of the study is to compare the performance of four deep learning architectures: VGG16, ResNet50, InceptionV3, and DenseNet201 in classifying Calamansi. Three types of optimizers were considered in the training including Adam, SGD, and RMSProp. Furthermore, data augmentation is applied to the datasets to reduce overfitting and increase variation in the training datasets because of limited data. With Adam as optimizer, classification accuracies resulted to 96.98%, 91.48%, 27.95% and 65.95% for ResNet50, VGG16, InceptionV3 and DenseNet201, respectively. Classification accuracies of 91.76% for ResNet50, 88.19% for VGG16, 25% for InceptionV3 and 39.84% for DenseNet201 where observed implementing the SGD optimizer. Lastly, 95.60%, 75.55%, 25% and 63.19% classification accuracies were obtained for ResNet50, VGG16, InceptionV3 and DensNet201, respectively when RMSProp was utilized as the optimizer. From the CNN architectures used, ResNet50 acquired the best result in classifying Calamansi. The model resulted in an overall accuracy of 96.98% utilizing Adam as optimizer. Based on the findings of the study, it was determined that Resnet50 architecture with Adam as optimizer may be utilized to categorize Calamansi fruit or the Philippine lime.
引用
收藏
页码:494 / 500
页数:7
相关论文
共 41 条
  • [1] Adigun J. O., AUTOMATED SYSTEM GRA
  • [2] Africa A. D. M., 2020, INT J EMERG TRENDS E, V8, P1845
  • [3] A powerful method for measuring fish size of small-scale fishery catches using ImageJ
    Andrialovanirina, Nicolas
    Ponton, Dominique
    Behivoke, Faustinato
    Mahafina, Jamal
    Leopold, Marc
    [J]. FISHERIES RESEARCH, 2020, 223
  • [4] Automated disease classification in (Selected) agricultural crops using transfer learning
    Aravind, Krishnaswamy Rangarajan
    Raja, Purushothaman
    [J]. AUTOMATIKA, 2020, 61 (02) : 260 - 272
  • [5] Mechatronic components in apple sorting machines with computer vision
    Baneh, Nesar Mohammadi
    Navid, Hosein
    Kafashan, Jalal
    [J]. JOURNAL OF FOOD MEASUREMENT AND CHARACTERIZATION, 2018, 12 (02) : 1135 - 1155
  • [6] Analysis of various optimizers on deep convolutional neural network model in the application of hyperspectral remote sensing image classification
    Bera, Somenath
    Shrivastava, Vimal K.
    [J]. INTERNATIONAL JOURNAL OF REMOTE SENSING, 2020, 41 (07) : 2664 - 2683
  • [7] Bhole Varsha, 2020, SIGITE '20: Proceedings of the 21st Annual Conference on Information Technology Education, P180, DOI 10.1145/3368308.3415370
  • [8] Bhole V., 2020, ICT for competitive strategies, V1, P759, DOI [10.1201/9781003052098-80, DOI 10.1201/9781003052098-80]
  • [9] Chowdhurry M.E., 2021, TOMATO LEAF DIS DETE
  • [10] Custodio E. F., 2021, 2021 IEEE 19 STUDENT