Recycling waste classification using optimized convolutional neural network

Mao, Wei‐Lung; Chen, Wei‐Chun; Wang, Chien-Tsung; Lin, Yun-Huin

doi:10.1016/j.resconrec.2020.105132

Cited by 177 publications

(78 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Conventional automatic sorting systems are based on different types of sensors (e.g., optical [28]- [30] and thermal techniques [31], [32]). Mao et al [33] proposed a classifier using a convolutional neural network to classify an RGB object image that included one waste item. Furthermore, DL-based algorithms using RGB and RGB-depth (RGBD) sensors have been used to detect and segment individual waste items from a densely cluttered pile [6], [22], [34]- [36].…”

Section: A Robotic Waste Sortermentioning

confidence: 99%

Robotic Waste Sorter With Agile Manipulation and Quickly Trainable Detector

Kiyokawa¹,

Katayama

Tatsuta

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Owing to human labor shortages, the automation of labor-intensive manual waste-sorting is needed. The goal of automating waste-sorting is to replace the human role of robust detection and agile manipulation of waste items with robots. To achieve this, we propose three methods. First, we provide a combined manipulation method using graspless push-and-drop and pick-and-release manipulation. Second, we provide a robotic system that can automatically collect object images to quickly train a deep neuralnetwork model. Third, we provide a method to mitigate the differences in the appearance of target objects from two scenes: one for dataset collection and the other for waste sorting in a recycling factory. If differences exist, the performance of a trained waste detector may decrease. We address differences in illumination and background by applying object scaling, histogram matching with histogram equalization, and background synthesis to the source target-object images. Via experiments in an indoor experimental workplace for waste-sorting, we confirm that the proposed methods enable quick collection of the training image sets for three classes of waste items (i.e., aluminum can, glass bottle, and plastic bottle) and detection with higher performance than the methods that do not consider the differences. We also confirm that the proposed method enables the robot quickly manipulate the objects.

show abstract

Section: A Robotic Waste Sortermentioning

confidence: 99%

Robotic Waste Sorter With Agile Manipulation and Quickly Trainable Detector

Kiyokawa¹,

Katayama

Tatsuta

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, this system performed poorly compared to the system based on extracting Scale Invariant Feature (SIFT) [52] and feeding it to the SVM classifier [43]. The authors in [53] proposed a waste sorting system for all types of materials. It is based on DenseNet-121 [54] deep learning architecture.…”

Section: Convolutional Neural Net Based Approachesmentioning

confidence: 99%

“…It is based on DenseNet-121 [54] deep learning architecture. The choice of DenseNet was motivated by the small size of the dataset [53]. In an attempt to improve the performance, data augmentation is employed by considering vertical, horizontal, and random 25° rotations.…”

Section: Convolutional Neural Net Based Approachesmentioning

confidence: 99%

Computer Vision based Polyethylene Terephthalate (PET) Sorting for Waste Recycling

Bchir¹,

Alghannam²,

Alsadhan³

et al. 2021

IJACSA

View full text Add to dashboard Cite

“…For example, intelligent classification of glass and metal in garbage bags by training convolutional neural network (CNN) [14]. Optimize the fully connected layer of CNN through genetic algorithm (GA) can improve the performance of waste detection [15]. An intelligent system of waste classification based on ResNet [16] can achieve accurate waste classification [17].…”

Section: Introductionmentioning

confidence: 99%

Skip-YOLO: Domestic Garbage Detection Using Deep Learning Method in Complex Multi-scenes

Zhao

Pan

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

In In the field of garbage intelligent identification, similar garbage are difficult to be effectively detected due to different kinds of characteristics. This paper proposes a Skip-YOLO model for garbage detection in real life through the visual analysis of feature mapping in different neural networks. First of all, the receptive field of the model is enlarged through the large-size convolution kernel, which enhanced the shallow information of images. Secondly, the high-dimensional feature mappings of garbage is extracted by dense convolutional blocks. The sensitivity of similar features in the same type of garbage is enhanced by strengthening the sharing of shallow low semantics and deep high semantics information. Finally, the multi-scale high-dimensional feature mappings is integrated and sent to the YOLO layer to predict the type and location of garbage. Experimental results show that compared with the YOLOv3, the overall detection precision is increased by 22.5%, and the average recall rate is increased by 18.6%. In qualitative comparison, it successfully detects domestic garbage in complex multi-scenes. In addition, our approach alleviates the overfitting problem of deep residual blocks. The application case of waste sorting production line is used to further highlight the model generalization performance of our method.

show abstract

Recycling waste classification using optimized convolutional neural network

Cited by 177 publications

References 16 publications

Robotic Waste Sorter With Agile Manipulation and Quickly Trainable Detector

Robotic Waste Sorter With Agile Manipulation and Quickly Trainable Detector

Computer Vision based Polyethylene Terephthalate (PET) Sorting for Waste Recycling

Skip-YOLO: Domestic Garbage Detection Using Deep Learning Method in Complex Multi-scenes

Contact Info

Product

Resources

About