An Improved CNN-Based Apple Appearance Quality Classification Method With Small Samples

Sun, Li; Liang, Kaibo; Song, Yanxing; Wang, Yuzhi

doi:10.1109/access.2021.3077567

Cited by 21 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Third, although the experimental environment was slightly different, the findings are compared to previous research (Frid-Adar et al (2018), Sajjad et al (2019), Sun et al (2021)). In the two previous studies that compared CNN classification performance with only nongenerated DA and nongenerated DA + DCGAN (Frid-Adar et al (2018), Sun et al (2021)), the discrimination accuracy with nongenerated DA + DCGAN was higher. The researchers used images in which the classification target accounts for a large proportion of the entire image.…”

Section: Discussionmentioning

confidence: 81%

See 1 more Smart Citation

Deep Anomaly Detection for Automotive Components by Oversampling

Yokocho,

Kawamura,

Nirasawa

2023

TQS

View full text Add to dashboard Cite

Training of deep neural networks (DNNs) requires large amounts of data. However, the automotive components that are the subject of this research have an extreme lack of defective product data due to rapid model changes and a low defective product rate during the manufacturing process. Additionally, the anomaly areas are negligible. Data augmentation (DA), which increases data by image transformations, is a method for solving data deficiency. Particularly, a deep convolutional generative adversarial network (DCGAN) is frequently employed in the medical industry. DA is shown to have an effect on not small anomalies but on images that are accounted by the classification target for a large percentage of the total image.Therefore, in the present study, we find the DA method that fits objects with very small anomaly areas. We use a method that combines existing methods and DCGAN because DCGAN only cannot be used in cases where target images are few. After increasing the data to some extent using existing methods for only the anomaly area, we increase it further using DCGAN and then paste the completed defect to the component images.The classification performance using DA, which modifies size, shape, and the like, based on the inspector's experience, yielded a recall value of 76.9%, whereas the performance using DA and DCGAN yielded a slightly lower recall value of 65.4%.

show abstract

Section: Discussionmentioning

confidence: 81%

“…An improved method based on CNN for apple appearance and quality classification with small samples was proposed by Sun et al (2021). Images of a single apple were utilized.…”

Section: Introductionmentioning

confidence: 99%

Deep Anomaly Detection for Automotive Components by Oversampling

Yokocho,

Kawamura,

Nirasawa

2023

TQS

View full text Add to dashboard Cite

show abstract

“…It has 50 layers for data processing. To avoid gradient dispersion/explosion and network degradation issues caused by excessively deep networks, ResNet50 has the advantage of using a jumping layer connection to a deep neural network [50].…”

Section: Architecture Of Resnet50mentioning

confidence: 99%

Brain cone beam computed tomography image analysis using ResNet50 for collateral circulation classification

Ali

Abdullah

Saad

et al. 2023

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">Treatment of stroke patients can be effectively carried out with the help of collateral circulation performance. Collateral circulation scoring as it is now used is dependent on visual inspection, which can lead to an inter- and <br /> intra-rater discrepancy. In this study, a collateral circulation classification using the ResNet50 was analyzed by using cone beam computed tomography (CBCT) images for the ischemic stroke patient. The remarkable performance of deep learning classification helps neuroradiologists with fast image classification. A pre-trained deep network ResNet50 was applied to extract robust features and learn the structure of CBCT images in their convolutional layers. Next, the classification layer of the ResNet50 was performed into binary classification as “good” and “poor” classes. The images were divided by 80:20 for training and testing. The empirical results support the claim that the application of ResNet50 offers consistent accuracy, sensitivity, and specificity values. The performance value of the classification accuracy was 76.79%. The deep learning approach was employed to unveil how biological image analysis could generate incredibly dependable and repeatable outcomes. The experiments performed on CBCT images evidenced that the proposed ResNet50 using convolutional neural network (CNN) architecture is indeed effective in classifying collateral circulation.</span>

show abstract

“…Accuracy/% LeNet-FC 98.9 AlexNet [20] 98.1 ResNet [21] 95.6 DCGAN [22] 92.5 SCNN [23] 94.7 PAMSGAN [24] 95.1…”

Section: Modelmentioning

confidence: 99%

A novel deep LeNet-5 convolutional neural network model for image recognition

Zhang

Lei

et al. 2022

COMSIS J

View full text Add to dashboard Cite

At present, the traditional machine learning methods and convolutional neural network (CNN) methods are mostly used in image recognition. The feature extraction process in traditional machine learning for image recognition is mostly executed by manual, and its generalization ability is not strong enough. The earliest convolutional neural network also has many defects, such as high hardware requirements, large training sample size, long training time, slow convergence speed and low accuracy. To solve the above problems, this paper proposes a novel deep LeNet-5 convolutional neural network model for image recognition. On the basis of Lenet-5 model with the guaranteed recognition rate, the network structure is simplified and the training speed is improved. Meanwhile, we modify the Logarithmic Rectified Linear Unit (L_ReLU) of the activation function. Finally, the experiments are carried out on the MINIST character library to verify the improved network structure. The recognition ability of the network structure in different parameter s is analyzed compared with the state-of-the-art recognition algorithms. In terms of the recognition rate, the proposed method has exceeded 98%. The results show that the accuracy of the proposed structure is significantly higher than that of the other recognition algorithms, which provides a new reference for the current image recognition.

show abstract

An Improved CNN-Based Apple Appearance Quality Classification Method With Small Samples

Cited by 21 publications

References 32 publications

Deep Anomaly Detection for Automotive Components by Oversampling

Deep Anomaly Detection for Automotive Components by Oversampling

Brain cone beam computed tomography image analysis using ResNet50 for collateral circulation classification

A novel deep LeNet-5 convolutional neural network model for image recognition

Contact Info

Product

Resources

About