A two-stage CNN-based hand-drawn electrical and electronic circuit component recognition system

Dey, Mrityunjoy; Mia, Shoif Md; Sarkar, Navonil; Bhattacharya, Abhishek; Roy, Sonali; Malakar, Samir; Sarkar, Ram

doi:10.1007/s00521-021-05964-1

Cited by 16 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is frequently used in the task of segmenting images. [1], [14], [20], [22], [23], [25], [30], [43], [45], [46] 6) Data Augmentation: The robustness of the predictions can be increased by supplementing the input data for training using modifications of hand-drawn diagrams, such as various styles, layouts, and amounts of noise.…”

Section: A Pre-processing Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An Overview of Hand-Drawn Diagram Recognition Methods and Applications

Agrawal,

Jagtap,

Kantipudi

2024

IEEE Access

View full text Add to dashboard Cite

Hand-drawn diagrams have been a standard visual communication tool in many disciplines, including architectural design, engineering, and education. The inherent diversity and absence of standardized formats of hand-drawn diagrams make it difficult to recognize them. As a result, there is an increasing need for efficient strategies and approaches for correctly identifying and interpreting hand-drawn diagrams. This research study comprehensively reviews hand-drawn diagram recognition (HDDR) techniques, emphasizing their importance and usefulness in numerous sectors. For the past ten years, articles from the Scopus database on HDDR have been extracted and reviewed. The study explores the approaches, steps, and benchmark datasets available to recognize hand-drawn diagrams. An attempt is made to get insights into the most recent state-of-the-art methodologies, their limits, and potential future advancement directions. This paper also suggests probable solutions to overcome the limitations and develop new techniques for efficiently and robustly recognizing hand-drawn diagrams. INDEX TERMSComputer vision, deep learning, hand-drawn diagram recognition, machine learning, sketches.

show abstract

Section: A Pre-processing Methodsmentioning

confidence: 99%

“…Priya A. K. [28] Need to create a benchmark database of manually drawn circuits. Electronic circuits Dey M. [30] Unability to effectively distinguish between highly similar elements, such as ammeter and voltmeter, can be decreased through the inclusion of novel strategies reflecting the local features.…”

Section: Wartegg Handdrawingsmentioning

confidence: 99%

An Overview of Hand-Drawn Diagram Recognition Methods and Applications

Agrawal,

Jagtap,

Kantipudi

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Results showed that the robustness and the accuracy of these CNN‐based methods are satisfactory [26, 40]. For example, Dey et al [9] designed a two‐stage CNN, which had a group‐level classification and a component‐level classification, to recognize the hand‐drawn circuit components and showed that the proposed method was able to achieve an accuracy of 97.3%. Similarly, Keerthi Priya et al [25] adopted the VGG16 architecture to classify electronic circuit symbols and achieved an accuracy of 99.2%.…”

Section: Introductionmentioning

confidence: 99%

Image‐based structural analysis for education purposes: A proof‐of‐concept study

Loong

Juan

Chang

2023

Comp Applic In Engineering

View full text Add to dashboard Cite

In civil engineering education, hand‐drawn sketches of structural systems are commonly used for response analysis. Based on these sketches, students may determine the response of the structures either by hand calculations or via some specialized software through the construction of finite‐element models. The former method is prone to errors, while the latter method could be time‐consuming for inexperienced students. It would be convenient if the information within the hand‐drawn sketches could be automatically converted into computer‐recognizable objects for further structural analysis. To address this issue, a novel method entitled Image‐based Structural Analysis (ISA) is proposed as a proof of concept to determine the response of a linear‐elastic structure directly from the image of a hand‐drawn sketch. A selective search algorithm and a deep convolutional neural network are adopted to detect relevant objects in the images. Based on the bounding boxes and the classes of the objects, a finite element model is constructed for further structural response analysis. This study demonstrates the proposed ISA method via several hand‐drawn beams under loadings. Results show that the proposed method, which consists of a combination of artificial intelligence and semantic rules, can produce, and analyze structural models from hand‐drawn sketches.

show abstract

“…Subsequently, with the development of deep learning, there have been studies integrating it into the recognition of hand-drawn electrical schematics [ 4 , 5 , 6 ]. The most recent papers can be traced back to 2015, when De et al [ 7 ] proposed a circuit image segmentation algorithm.…”

Section: Introductionmentioning

confidence: 99%

Parsing Netlists of Integrated Circuits from Images via Graph Attention Network

Hu,

Zhan,

Tong

2023

Sensors

View full text Add to dashboard Cite

A massive number of paper documents that include important information such as circuit schematics can be converted into digital documents by optical sensors like scanners or digital cameras. However, extracting the netlists of analog circuits from digital documents is an exceptionally challenging task. This process aids enterprises in digitizing paper-based circuit diagrams, enabling the reuse of analog circuit designs and the automatic generation of datasets required for intelligent design models in this domain. This paper introduces a bottom-up graph encoding model aimed at automatically parsing the circuit topology of analog integrated circuits from images. The model comprises an improved electronic component detection network based on the Swin Transformer, an algorithm for component port localization, and a graph encoding model. The objective of the detection network is to accurately identify component positions and types, followed by automatic dataset generation through port localization, and finally, utilizing the graph encoding model to predict potential connections between circuit components. To validate the model’s performance, we annotated an electronic component detection dataset and a circuit diagram dataset, comprising 1200 and 3552 training samples, respectively. Detailed experimentation results demonstrate the superiority of our proposed enhanced algorithm over comparative algorithms across custom and public datasets. Furthermore, our proposed port localization algorithm significantly accelerates the annotation speed of circuit diagram datasets.

show abstract

A two-stage CNN-based hand-drawn electrical and electronic circuit component recognition system

Cited by 16 publications

References 25 publications

An Overview of Hand-Drawn Diagram Recognition Methods and Applications

An Overview of Hand-Drawn Diagram Recognition Methods and Applications

Image‐based structural analysis for education purposes: A proof‐of‐concept study

Parsing Netlists of Integrated Circuits from Images via Graph Attention Network

Contact Info

Product

Resources

About