A combination model based on transfer learning for waste classification

Huang, Guang-Li; He, Jing; Xu, Zenglin; Huang, Guangyan

doi:10.1002/cpe.5751

Cited by 59 publications

(16 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The unique property of DenseNet 169 is the connection between each layer in a Dense block and all the subsequent layers in that block. Researchers have used DenseNet to solve the classification tasks related to waste classification [ 38 ], multiple sclerosis classification [ 39 ], monocular depth estimation [ 40 ] and lung nodule classification [ 41 ].…”

Section: Preliminariesmentioning

confidence: 99%

A stacked ensemble for the detection of COVID-19 with high recall and accuracy

Jangam

Annavarapu

2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

The main challenges for the automatic detection of the coronavirus disease (COVID-19) from computed tomography (CT) scans of an individual are: a lack of large datasets, ambiguity in the characteristics of COVID-19 and the detection techniques having low sensitivity (or recall). Hence, developing diagnostic techniques with high recall and automatic feature extraction using the available data are crucial for controlling the spread of COVID-19. This paper proposes a novel stacked ensemble capable of detecting COVID-19 from a patient’s chest CT scans with high recall and accuracy. A systematic approach for designing a stacked ensemble from pre-trained computer vision models using transfer learning (TL) is presented. A novel diversity measure that results in the stacked ensemble with high recall and accuracy is proposed. The stacked ensemble proposed in this paper considers four pre-trained computer vision models: the visual geometry group (VGG)-19, residual network (ResNet)-101, densely connected convolutional network (DenseNet)-169 and wide residual network (WideResNet)-50-2. The proposed model was trained and evaluated with three different chest CT scans. As recall is more important than precision, the trade-offs between recall and precision were explored in relevance to COVID-19. The optimal recommended threshold values were found for each dataset.

show abstract

Section: Preliminariesmentioning

confidence: 99%

A stacked ensemble for the detection of COVID-19 with high recall and accuracy

Jangam

Annavarapu

2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

show abstract

“…One of the most important datasets is the one gathered during the development of TrashNet [ 7 ]. This dataset has been the basis for many other studies [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ]. The TrashNet dataset has been further extended by adding the compost class [ 10 ].…”

Section: Related Workmentioning

confidence: 99%

“…In this framework, RecycleNet [ 8 ] improved dramatically the performance obtained on the TrashNet dataset. RecycleNet extends a 121 layers DenseNet [ 22 ] with the use of skip connections via concatenation on Models with many layers combined with application of transfer learning.…”

Section: Related Workmentioning

confidence: 99%

“…For example, waste pickers of the most important landfills in Brazil have been, or still are, subject to many diseases, due to working conditions: mainly osteo-muscular disorders, but also viral infections and emotional vulnerabilities [ 6 ]. Therefore, current waste recognition systems [ 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ] as well as current corpora for training [ 7 , 10 ] focus on the activity of waste sorting in waste management plants. Solid waste runs on conveyor belts and, then, it is sorted into the different types.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Data Augmentation Using Background Replacement for Automated Sorting of Littered Waste

2021

View full text Add to dashboard Cite

The introduction of sophisticated waste treatment plants is making the process of trash sorting and recycling more and more effective and eco-friendly. Studies on Automated Waste Sorting (AWS) are greatly contributing to making the whole recycling process more efficient. However, a relevant issue, which remains unsolved, is how to deal with the large amount of waste that is littered in the environment instead of being collected properly. In this paper, we introduce BackRep: a method for building waste recognizers that can be used for identifying and sorting littered waste directly where it is found. BackRep consists of a data-augmentation procedure, which expands existing datasets by cropping solid waste in images taken on a uniform (white) background and superimposing it on more realistic backgrounds. For our purpose, realistic backgrounds are those representing places where solid waste is usually littered. To experiment with our data-augmentation procedure, we produced a new dataset in realistic settings. We observed that waste recognizers trained on augmented data actually outperform those trained on existing datasets. Hence, our data-augmentation procedure seems a viable approach to support the development of waste recognizers for urban and wild environments.

show abstract

“…Convolutional neural network (CNN) is the most extensively utilized deep learning technique for object classification and detection. [20][21][22] In the field of vehicle detection, Chen et al 18 presented a technique based on sliding-windows and CNNs. The idea of this method is to reduplicate the convolution layers of the network at different scales, enabling the deep learning model to detect vehicles at different scales.…”

mentioning

confidence: 99%

SI‐EDTL: Swarm intelligence ensemble deep transfer learning for multiple vehicle detection in UAV images

Darehnaei

Shokouhifar

Yazdanjouei

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

This article proposes a swarm intelligence ensemble deep transfer learning (named SI-EDTL) for multiple vehicle detection in unmanned aerial vehicle (UAV) images.This method is based on Faster regional-based convolutional neural networks (Faster R-CNN), in which, a set of region proposals are extracted using region proposal network (RPN), and then, CNN is used to mine highly descriptive features of these windows to classify regions. We use three Faster R-CNNs as feature extractors (InceptionV3, ResNet50, and GoogLeNet) that have already pre-trained on Ima-geNet data, combined with five transfer classifiers (KNN, SVM, MLP, C4.5 Decision Tree, and Naïve Bayes). As a result, 15 different base learners are trained through deep transfer learning on a UAV dataset to classify the region proposals into multiple vehicles (car, van, truck, and bus). We combine these 15 base learners through a weighted averaging aggregation into four vehicle classes or no vehicle (background). Hyperparameters of the ensemble model are tuned using whale optimization algorithm, to achieve the best trade-off between total accuracy, precision, and recall. The proposed SI-EDTL model has been successfully developed using parallel processing in MATLAB R2020b. Experimental results on AU-AIR dataset of UAV images demonstrate the superiority of the SI-EDTL model against existing techniques.

show abstract

A combination model based on transfer learning for waste classification

Cited by 59 publications

References 32 publications

A stacked ensemble for the detection of COVID-19 with high recall and accuracy

A stacked ensemble for the detection of COVID-19 with high recall and accuracy

Data Augmentation Using Background Replacement for Automated Sorting of Littered Waste

SI‐EDTL: Swarm intelligence ensemble deep transfer learning for multiple vehicle detection in UAV images

Contact Info

Product

Resources

About