Corn Ear Quality Recognition Based on DCGAN Data Enhancement and Transfer Learning

Shi, Benzheng; Zhou, Xiaolei; Qin, Zhikang; Sun, Luna; Xu, Yan

doi:10.1145/3462676.3462686

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Where G is the generative network, D is the discriminant network, 𝐺(𝑧) is the image obtained by the generator, 𝐷(𝐺(𝑧)) is the discrimination probability of the generated image, and 𝐷(𝑥) is the discrimination probability of the real image [15].…”

Section: B Improved Dcgan Loss Functionmentioning

confidence: 99%

“…The dataset images are then randomly divided into two groups: 80% of the dataset is used for parameter learning and network training, whereas the other 20% is used to test the generalization and recognition ability of the model, and the two datasets do not intersect each other. (15,30), (13,48), (19,38), (17,59), (14,81), (17,75), (24,55), (19,100), (14,144). The distribution is shown in Table 2.…”

Section: A Preparation Of Adhesive Structures Defect Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Defect Identification of Adhesive Structure Based on DCGAN and YOLOv5

et al. 2022

View full text Add to dashboard Cite

To overcome the problem of small defect samples and the imbalanced distribution of defect categories during adhesive structure defect detection, a defect identification approach based on DCGAN and YOLOv5 is proposed. The above problems are solved by fine-tuning the structure and the loss function of the DCGAN, generating high-quality defect images, and expanding the dataset of adhesive structure defects. Generally, using the EIOU loss function in the YOLOv5 network allows the network to converge faster during training and improve the recognition effect. In this article, it is illustrated by comparing with the GIOU loss function, that utilizing the EIOU loss function increases the mAP value by 3.9% and the recall by 10.5%, but the precision decreases. To solve this problem, the feature extraction capability of the network is enhanced by incorporating the CBAM after the C3 module in the YOLOv5 network. Interestingly, the mAP, precision, and recall of the optimized YOLOv5 algorithm are improved to 78.6%, 77.2%, and 76%, respectively, where the precision compared to the original model improved by 10.6%. This study has demonstrated that the improved YOLOv5 model can effectively detect adhesive structure defects, which provides defect identification and control theoretical research and technical support.

show abstract

Section: B Improved Dcgan Loss Functionmentioning

confidence: 99%

Section: A Preparation Of Adhesive Structures Defect Datasetmentioning

confidence: 99%

Defect Identification of Adhesive Structure Based on DCGAN and YOLOv5

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In recent years, the research of plant phenotypes based on image semantic segmentation has been playing an important role. However, the current research mainly focuses on the classification of abnormal regions of the whole ear, and has not achieved the fine segmentation of abnormal regions of the single ear [5,6]. Most researchers extracted the image colour, texture, shape and other features [7,8], and further used the image segmentation algorithm for implementation.…”

Section: Related Workmentioning

confidence: 99%

Abnormal area identification of corn ear based on semi‐supervised learning

Wei

Wang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Screening corn ear is the key link in the breeding process of new varieties. But manual testing is difficult to measure the proportion of abnormal area. Meanwhile, the abnormal areas are mainly caused by mildew, moth and mechanical collision. A new refined semantic segmentation model was proposed based on the semi‐supervised learning method of generating antagonistic networks (GAN). Besides k‐means algorithm was used to remove a large amount of background information for data preprocessing. By using feature fusion and weighted loss function the model performance was improved. The introduction of transfer learning accelerated model convergence. Through the high‐throughput corn ear collection system, 1448 ear images (including abnormal conditions such as mildew, moth and mechanical damage) were collected and labelled. The proposed method was tested on real corn ear images with an accuracy of 0.950, mean precision of 0.933, mean IoU of 0.884, and FwIoU of 0.908. Experimental results show that the proposed method has better performance than general networks.

show abstract

“…Custom CNN models provided improved results compared to transfer learning and they gives an accuracy of 97.37% ( 20 ). The Dense Attentive Circular Network is used for malaria cell detection and achieved results are compared to the transfer learning models such as DPN92 and DenseNet121.This method provided an accuracy of 97.47 and 87.88% on DenseNet121 and DPN-92, respectively ( 21 ). The pre-trained networks such as ResNet50, AlexNet, and VGG19 are used for malaria classification and provide an accuracy of 93.88, 96.33, and 93.72%, respectively.…”

Section: Related Workmentioning

confidence: 99%

An optimized features selection approach based on Manta Ray Foraging Optimization (MRFO) method for parasite malaria classification

Amin

Sharif

Mallah

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

Malaria is a serious and lethal disease that has been reported by the World Health Organization (WHO), with an estimated 219 million new cases and 435,000 deaths globally. The most frequent malaria detection method relies mainly on the specialists who examine the samples under a microscope. Therefore, a computerized malaria diagnosis system is required. In this article, malaria cell segmentation and classification methods are proposed. The malaria cells are segmented using a color-based k-mean clustering approach on the selected number of clusters. After segmentation, deep features are extracted using pre-trained models such as efficient-net-b0 and shuffle-net, and the best features are selected using the Manta-Ray Foraging Optimization (MRFO) method. Two experiments are performed for classification using 10-fold cross-validation, the first experiment is based on the best features selected from the pre-trained models individually, while the second experiment is performed based on the selection of best features from the fusion of extracted features using both pre-trained models. The proposed method provided an accuracy of 99.2% for classification using the linear kernel of the SVM classifier. An empirical study demonstrates that the fused features vector results are better as compared to the individual best-selected features vector and the existing latest methods published so far.

show abstract

Corn Ear Quality Recognition Based on DCGAN Data Enhancement and Transfer Learning

Cited by 3 publications

References 0 publications

Defect Identification of Adhesive Structure Based on DCGAN and YOLOv5

Defect Identification of Adhesive Structure Based on DCGAN and YOLOv5

Abnormal area identification of corn ear based on semi‐supervised learning

An optimized features selection approach based on Manta Ray Foraging Optimization (MRFO) method for parasite malaria classification

Contact Info

Product

Resources

About