PCB-METAL: A PCB Image Dataset for Advanced Computer Vision Machine Learning Component Analysis

Mahalingam, Gayathri; Gay, Kevin Marshall; Ricanek, Karl

doi:10.23919/mva.2019.8757928

Cited by 29 publications

(21 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Typically, a conventional image contains three channels: Red, Green and Blue (RGB). Several studies use hyperspectral images that contain several channels to form reflectance spectra [94][95][96]. These images provide significantly more information than conventional RGB images and could enable higher classification accuracies.…”

Section: Machine Visionmentioning

confidence: 99%

Dismantling of Printed Circuit Boards Enabling Electronic Components Sorting and Their Subsequent Treatment Open Improved Elemental Sustainability Opportunities

et al. 2021

View full text Add to dashboard Cite

This critical review focuses on advanced recycling strategies to enable or increase recovery of chemical elements present in waste printed circuit boards (WPCBs). Conventional recycling involves manual removal of high value electronic components (ECs), followed by raw crushing of WPCBs, to recover main elements (by weight or value). All other elements remain unrecovered and end up highly diluted in post-processing wastes or ashes. To retrieve these elements, it is necessary to enrich the waste streams, which requires a change of paradigm in WPCB treatment: the disassembly of WPCBs combined with the sorting of ECs. This allows ECs to be separated by composition and to drastically increase chemical element concentration, thus making their recovery economically viable. In this report, we critically review state-of-the-art processes that dismantle and sort ECs, including some unpublished foresight from our laboratory work, which could be implemented in a recycling plant. We then identify research, business opportunities and associated advanced retrieval methods for those elements that can therefore be recovered, such as refractory metals (Ta, Nb, W, Mo), gallium, or lanthanides, or those, such as the platinum group elements, that can be recovered in a more environmentally friendly way than pyrometallurgy. The recovery methods can be directly tuned and adapted to the corresponding stream.

show abstract

Section: Machine Visionmentioning

confidence: 99%

Dismantling of Printed Circuit Boards Enabling Electronic Components Sorting and Their Subsequent Treatment Open Improved Elemental Sustainability Opportunities

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Moreover, many of the available datasets are intended and formatted for a specific study, so they are difficult to combine effectively and re-purpose for a different study. For example, FICS-PCB [16] consists of 9900 images taken with a Digital Single-Lens Reflex (DSLR) camera and optical microscope with six component classes annotated, and PCB-Metal [17] consists of only 1000 images taken with a DSLR camera with four component classes annotated. While these datasets provide sufficient information to experiment with various algorithms, much work is needed to increase and maintain data to be representative of as many components and their variations as possible.…”

Section: The Need For a Datasetmentioning

confidence: 99%

Why Is Deep Learning Challenging for Printed Circuit Board (PCB) Component Recognition and How Can We Address It?

et al. 2021

View full text Add to dashboard Cite

In this paper, we present the need for specialized artificial intelligence (AI) for counterfeit and defect detection of PCB components. Popular computer vision object detection techniques are not sufficient for such dense, low inter-class/high intra-class variation, and limited-data hardware assurance scenarios in which accuracy is paramount. Hence, we explored the limitations of existing object detection methodologies, such as region based convolutional neural networks (RCNNs) and single shot detectors (SSDs), and compared them with our proposed method, the electronic component localization and detection network (ECLAD-Net). The results indicate that, of the compared methods, ECLAD-Net demonstrated the highest performance, with a precision of 87.2% and a recall of 98.9%. Though ECLAD-Net demonstrated decent performance, there is still much progress and collaboration needed from the hardware assurance, computer vision, and deep learning communities for automated, accurate, and scalable PCB assurance.

show abstract

“…Preparing a representative dataset for Logo Classification in PCBs is difficult as mentioned. The FICS-PCB Dataset [10], although improving on other publicly available sets [24,25], has a limited number of PCBs, and hence a limited number of logos on them. To better train our classifier, we needed to add more images to the dataset.…”

Section: A the Need For Logo Augmentationmentioning

confidence: 99%

Logo Classification and Data Augmentation Techniques for PCB Assurance and Counterfeit Detection

Sathiaseelan

Paradis

Rai

et al. 2021

International Symposium for Testing and Failure Analysis

View full text Add to dashboard Cite

In this manuscript, we present our work on Logo classification in PCBs for Hardware assurance purposes. Identifying and classifying logos have important uses for text detection, component authentication and counterfeit detection. Since PCB assurance faces the lack of a representative dataset for classification and detection tasks, we collect different variants of logos from PCBs and present data augmentation techniques to create the necessary data to perform machine learning. In addition to exploring the challenges for image classification tasks in PCBs, we present experiments using Random Forest classifiers, Bag of Visual Words (BoVW) using SIFT and ORB Fully Connected Neural Networks (FCN) and Convolutional Neural Network (CNN) architectures. We present results and also a discussion on the edge cases where our algorithms fail including the potential for future work in PCB logo detection. The code for the algorithms along with the dataset that includes 18 classes of logos with 14000+ images is provided at this link: https://www.trusthub.org/#/data Index Terms—AutoBoM, Logo classification, Data augmentation, Bill of materials, PCB Assurance, Hardware Assurance, Counterfeit avoidance

show abstract

PCB-METAL: A PCB Image Dataset for Advanced Computer Vision Machine Learning Component Analysis

Cited by 29 publications

References 8 publications

Dismantling of Printed Circuit Boards Enabling Electronic Components Sorting and Their Subsequent Treatment Open Improved Elemental Sustainability Opportunities

Dismantling of Printed Circuit Boards Enabling Electronic Components Sorting and Their Subsequent Treatment Open Improved Elemental Sustainability Opportunities

Why Is Deep Learning Challenging for Printed Circuit Board (PCB) Component Recognition and How Can We Address It?

Logo Classification and Data Augmentation Techniques for PCB Assurance and Counterfeit Detection

Contact Info

Product

Resources

About