Page 8 - Read Online

Chazhoor et al. Intell Robot 2022;2:1-19 https://dx.doi.org/10.20517/ir.2021.15 Page 3

Figure 1. Types of plastic, its resin code and everyday examples of plastics. PETE: Polyethylene terephthalate; HDPE: high-density
polyethylene; PVC: polyvinyl chloride; LDPE: low-density polyethylene; PP: polypropylene, PS: polystyrene.

training time of a CNN by pre-training the model using benchmark datasets such as ImageNet.

Bobulski et al. proposed an end-to-end system with a micro-computer embedded with the vision to sort
[15]
the PETE types of plastics in the WaDaBa dataset. The authors introduced data augmentation, which
reduced the number of parameters but exponentially increased the number of samples, increasing the
training time. Bobulski et al. also proposed to classify distinct plastic categories based on a gradient
[16]
feature vector. Agarwal et al. presented Siamese and triplet loss neural networks to classify the WaDaBa
[17]
dataset and succeeded with very high accuracy. However, this method requires a significant amount of time
[18]
for training the neural networks. Chazhoor et al. Anthony utilised transfer learning to compare the three
most often used architectures (ResNeXt, Resnet-50-50 and AlexNet) on the WaDaBa dataset to select the
optimal model; however, the K-fold cross validation technique was not applied; as a result, testing accuracy
would vary widely.

The aim of the paper is to provide researchers with benchmark accuracies and the average time required to
train on the WaDaBa dataset using the latest CNN models utilising cross-validation to categorise a range of
plastics into their appropriate resin types. An unbiased and concrete set of parameters has been set to
[19]
evaluate the Dataset to compare the models fairly . This benchmark work will assist in gaining an
impartial view of numerous recent CNN models applied to the WaDaBa dataset, establishing a baseline for
[22]
[20]
[21]
future research. The models used in this paper are AlexNet , Resnet-50 , ResNeXt , SqueezeNet ,
[23]
MobileNet_v2 and DenseNet .
[25]
[24]