Maturity Classification of “Hupingzao” Jujubes with an Imbalanced Dataset Based on Improved MobileNet V2

Abstract

Fruits with various maturity levels coexist among the harvested jujubes, and have different tastes and uses. Manual grading has a low efficiency and a strong subjectivity. The number of “Hupingzao” jujubes between different maturity levels is unbalanced, which affects the performance of the classifier. To solve the above issue, the class balance loss (CB) was used to improve the MobileNet V2 network, and a transfer learning strategy was used to train the model. The model was optimized based on the selection of an optimizer and learning rate. The model achieved the best classification results using the AdamW optimizer and a learning rate of 0.0001. The application of transfer learning and class balance loss improved the model’s performance. The precision was 96.800~100.000%, the recall was 95.833~100.000%, and the F1 score was 0.963~1.000. To compare the CB-MobileNet V2 performance, the CB-AlexNet, CB-GoogLeNet, CB-ShuffleNet, CB-Inception V3, CB-ResNet 50, and CB-VGG 16 with transfer learning were used to build classification models. Achieving a validation accuracy of 99.058%, and a validation loss value of 0.055, the CB-MobileNet V2 model showed a better overall performance compared with other models. The maturity detection system of “Hupingzao” jujubes was developed to test the model. The testing accuracy of the CB-MobileNet V2 model was 99.294%. The research indicates that the CB-MobileNet V2 model improves the performance of maturity classification, and provides a theoretical basis for intelligent classification of the quality of “Hupingzao” jujubes.

Untitled

“…Mobile phones, laptops, and desktop computers may all access the Internet with the Jetson Nano [50][51][52]. Mobile Nets are built on a simplified design that creates lightweight deep neural networks using depth-wise separable convolutions [56][57][58][59][60][61][62][63]. In comparison to MobileNetV1, MobileNetV2 makes improvements to achieve higher accuracy with less input parameters and calculations [30,[56][57][58][59][60][61][62][63].…”

“…Mobile Nets are built on a simplified design that creates lightweight deep neural networks using depth-wise separable convolutions [56][57][58][59][60][61][62][63]. In comparison to MobileNetV1, MobileNetV2 makes improvements to achieve higher accuracy with less input parameters and calculations [30,[56][57][58][59][60][61][62][63]. Using depth-wise separable convolutions with linear bottlenecks and inverted residual blocks (shortcut connections between bottlenecks) we primarily introduce the key features of MobileNetV2, optimize the loss function, and use the improved model architecture from the Face Net model to illustrate MobileNetV2 [62].…”

Smart Attendance System based on improved Facial Recognition

“…5 shows the process of identifying and processing data. MobileNetV2 [37][38][39][40][41][42][43] improves over MobileNetV1 [26,44] to gain higher accuracy with fewer parameters and fewer calculations. In this section, we mainly introduce the core features of the MobileNetV2 to be used, the optimization of the loss function and utilizes the improved model architecture from FaceNet model.…”

Evaluation of maturity and moisture content in fresh jujube using fractal theory and tissue images

“…It has high nutritional value and medicinal value, and is one of the most popular fruits for consumers (Wang et al, 2022). The maturity and moisture content directly affect the taste and fruit quality of jujube (Fu et al, 2021;Sun et al, 2022;Wang et al, 2020), therefore, the researches on jujube maturity and moisture content detection have important significance. Many researchers have studied the detection methods of moisture content and maturity of jujube by selecting appropriate input data and building suitable predictive models.…”

“…Mahmood et al (2022) successfully divided jujubes into three categories (unripe, ripe, and over-ripe) based on the convolutional neural network (CNN) according to maturity, which provided theoretical support for the automatic classification system of jujube. Sun et al (2022) used the class balance loss (CB) to improve the MobileNetV2 network, and used A transfer learning strategy to train the model. The results showed that the CB-MobileNet V2 model improved the performance of jujube maturity classification.…”