Simultaneous mass estimation and class classification of scrap metals using deep learning

Díaz‐Romero, Dillam Jossue; Eynde, Simon Van den; Sterkens, Wouter; Engelen, Bart; Zaplana, Isiah; Dewulf, Wim; Goedemé, Toon; Peeters, Jef

doi:10.1016/j.resconrec.2022.106272

Cited by 19 publications

(7 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6 shows the regression results used for mass estimation of all test patterns. Previous research has shown that the density and, thus, the object's mass can be more accurately estimated if the object class is known, i.e., the network can learn the density per object type [18]. In this case, the mass estimation shows a trend by class with an 𝑅 2 of 0.75, RMSE of 1.39, and an MAE of 0.81 with an IoU of above 50%, as shown in Table III and Fig.…”

Section: Resultsmentioning

confidence: 99%

“…ONE CPU: AMD Ryzen Thread Ripper 3990x 64-core with 128 GB DDR4 RDIMM memory. 16 bits depth image requires a pre-processing step to convert the images to 8 bits [18]. The first step is to define an area of interest (AOI) that extends from the surface where the object is placed to the highest depth intensity on the sample.…”

Section: Methodsmentioning

confidence: 99%

“…However, the method did not allow the system to be used in real-world conditions since a fixed background and calibration are required for object detection. In 2022, the use of a convolutional neural network (CNN) for classification and a backpropagation neural network (BPNN) for mass estimation using up to 24 features extracted from depth images of metal scrap pieces was proposed [18]. However, this method does not use RGB images, which limits the detection and classification accuracy.…”

Section: A Related Work For Mass Estimation In Recyclingmentioning

confidence: 99%

“…The sub-loss functions 𝐿 CLs , 𝐿 bbox and 𝐿 mask are identical to those defined in [19], [21]. The L mass is the Mean Absolute Error (MAE or 𝐿 1 -Loss) and has been used in previous works for mass estimation [18].…”

Section: F Implementation Detailsmentioning

confidence: 99%

See 3 more Smart Citations

Enhanced Plastic Recycling using RGB+Depth Fusion with MassFaster and MassMask R-CNN

Díaz‐Romero

Zaplana

Eynde

et al. 2022

2022 Fourth International Conference on Transdisciplinary AI (TransAI)

Self Cite

View full text Add to dashboard Cite

The rapid increase in waste generation from electrical and electronic equipment (WEEE) has created the need for more advanced sensor-based systems to sort this complex type of waste. Therefore, this study proposes a method for object detection, instance segmentation, and mass estimation of plastics and contaminants using the fusion of RGB and depth (D) images. The methodology is based on the Faster and Mask R-CNN with an extra head for the mass estimation. In addition, a pre-processing method to enhance the depth image (ED) is proposed. To evaluate the data fusion and pre-processing method, two data sets of plastics and impurities were created containing images with and without overlapping samples. The first data set contains 174 RGB images and depth (D) maps of 3146 samples, excluding their mass value, while the second data set contains 42 RGB and D images of 766 pieces together with their mass. The first and second data sets were used to evaluate the performance of Mask and Faster R-CNN. Further, the second data set was used to evaluate the network's performance with the additional head for mass estimation.The proposed method achieved 0.75 𝑹 𝟐 , 1.39 RMSE, and 0.81 MAE with an IoU greater than 50% using the network Resnet50_FPN_RGBED. Hence, it can be concluded that the presented method can distinguish plastics from other materials with good accuracy. Furthermore, the mass of each detected particle can be estimated individually, which is of great relevance for the recycling sector. Knowing the mass distribution and the percentage of contaminants in a waste stream of mixed plastics can be valuable for adjusting the parameters of upstream and downstream sorting processes.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: A Related Work For Mass Estimation In Recyclingmentioning

confidence: 99%

Section: F Implementation Detailsmentioning

confidence: 99%

See 2 more Smart Citations

Enhanced Plastic Recycling using RGB+Depth Fusion with MassFaster and MassMask R-CNN

Díaz‐Romero

Zaplana

Eynde

et al. 2022

2022 Fourth International Conference on Transdisciplinary AI (TransAI)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this paper, we report a first numerical study of such a network (see Sections III-B4 and IV-D). While both mass estimation [29]- [32] and material recognition [33]- [36] from images have been proposed to improve robotic manipulation, in this paper they are relevant for performing autonomous resources mapping and quantification (see Section 4 of [11]): the former quantifies the amount of material, the latter specifies the type.…”

Section: Computer Vision For Resources Monitoringmentioning

confidence: 99%

Thermodynamical Material Networks for Modeling, Planning, and Control of Circular Material Flows

Zocco

Sopasakis

Smyth

et al. 2023

International Journal of Sustainable Engineering

View full text Add to dashboard Cite

The linear take-make-dispose paradigm at the foundations of our traditional economy is proving to be unsustainable due to waste pollution and material supply uncertainties. Hence, increasing the circularity of material flows is necessary. In this paper, we make a step towards circular healthcare by developing several vision systems targeting three main circular economy tasks: resources mapping and quantification, waste sorting, and disassembly. The performance of our systems demonstrates that representation-learning vision can improve the recovery chain, where autonomous systems are key enablers due to the contamination risks. We also published two fully-annotated datasets for image segmentation and for key-point tracking in disassembly operations of inhalers and glucose meters. The datasets and source code are publicly available 1 .

show abstract

Prediction of Converter Tapping Weight Based on Principal Component Analysis–Whale Optimization Algorithm–Backpropagation Algorithm Neural Network Model

Xue,

Xu,

Bao

et al. 2024

steel research int.

View full text Add to dashboard Cite

Tapping weight is an important parameter in converter blowing process, wherein precisely predicting the quantity of steel required in an alloy baking converter can effectively guide the requirement of alloy ingredients. In practical production, the main approach is empirical estimation, despite its low accuracy. Employing a general neural network model for prediction requires to gather the converter blowing parameters and endpoint temperature measurement sampling parameters as the model inputs. However, this data cannot be obtained until the blowing process reaches its endpoint, rendering it impractical for alloy batching that requires advance preparation for baking. In this study, a principal component analysis–whale optimization algorithm–backpropagation algorithm (PCA–WOA–BP) neural network tapping prediction model is developed using raw material parameters available before alloy baking as input. This model is integrated into the intelligent alloy reduction model of a factory. The model achieves a regression coefficient R2 of 0.823, with 98.53% of furnaces having a prediction error of less than 2 t. The tapping weight of 20 consecutive heats in actual production is predicted; the error range is less than 80% within 1 t, and the converter tapping weight is predicted accurately.

show abstract

Simultaneous mass estimation and class classification of scrap metals using deep learning

Cited by 19 publications

References 41 publications

Enhanced Plastic Recycling using RGB+Depth Fusion with MassFaster and MassMask R-CNN

Enhanced Plastic Recycling using RGB+Depth Fusion with MassFaster and MassMask R-CNN

Thermodynamical Material Networks for Modeling, Planning, and Control of Circular Material Flows

Prediction of Converter Tapping Weight Based on Principal Component Analysis–Whale Optimization Algorithm–Backpropagation Algorithm Neural Network Model

Contact Info

Product

Resources

About