Automatic Extraction of Indoor Spatial Information from Floor Plan Image: A Patch-Based Deep Learning Methodology Application on Large-Scale Complex Buildings

Kim, Hyun-Jung; Kim, Seong-Yong; Yu, Kiyun

doi:10.3390/ijgi10120828

Cited by 11 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These were used to prepare a set of rules describing the task-specific context and were reported to improve the performance achieved. Two different approaches to floor plan analysis and environment model generation were presented by Kim et al [121,125]. The first article focused on the problem of non-existing standardization in the floor plan format.…”

Section: Segmentationmentioning

confidence: 99%

Methods and Applications of Space Understanding in Indoor Environment—A Decade Survey

Pokuciński,

Mrozek

2024

Applied Sciences

View full text Add to dashboard Cite

The demand for digitizing manufacturing and controlling processes has been steadily increasing in recent years. Digitization relies on different techniques and equipment, which produces various data types and further influences the process of space understanding and area recognition. This paper provides an updated view of these data structures and high-level categories of techniques and methods leading to indoor environment segmentation and the discovery of its semantic meaning. To achieve this, we followed the Systematic Literature Review (SLR) methodology and covered a wide range of solutions, from floor plan understanding through 3D model reconstruction and scene recognition to indoor navigation. Based on the obtained SLR results, we identified three different taxonomies (the taxonomy of underlying data type, of performed analysis process, and of accomplished task), which constitute different perspectives we can adopt to study the existing works in the field of space understanding. Our investigations clearly show that the progress of works in this field is accelerating, leading to more sophisticated techniques that rely on multidimensional structures and complex representations, while the processing itself has become focused on artificial intelligence-based methods.

show abstract

Section: Segmentationmentioning

confidence: 99%

Methods and Applications of Space Understanding in Indoor Environment—A Decade Survey

Pokuciński,

Mrozek

2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Amongst these applications, there has been a considerable research focus on P&IDs (Rahul et al 2019;Sinha et al 2019;Yu et al 2019;Mani et al 2020;Gao et al 2020;Elyan et al 2020a;Moreno-García et al 2020;Jamieson et al 2020;Nurminen et al 2020;Paliwal et al 2021a;Moon et al 2021;Kim et al 2021b;Stinner et al 2021;Paliwal et al 2021b;Toral et al 2021;Bhanbhro et al 2022;Hantach et al 2021). Another research area is architecture diagram digitisation (Ziran and Marinai 2018;Zhao et al 2020;Rezvanifar et al 2020;Kim et al 2021a;Renton et al 2021;Jakubik et al 2022). Deep learning methods were also applied to technical drawings (Nguyen et al 2021), construction drawings (Faltin et al 2022) engineering documents (Francois et al 2022) and engineering drawings (Sarkar et al 2022;Scheibel et al 2021;Haar et al 2023).…”

Section: Application Domainsmentioning

confidence: 99%

“…Deep learning has also been recently applied for the digitisation of architecture diagrams (Ziran and Marinai 2018;Zhao et al 2020;Rezvanifar et al 2020;Kim et al 2021a;Renton et al 2021;Jakubik et al 2022). These present similar challenges to engineering diagrams, such as various semantically equivalent symbol representations (Rezvanifar et al 2020), relatively small objects (Kim et al 2021a) and the presence of occlusion and clutter (Rezvanifar et al 2020). One example is the work by Zhao et al (2020), which proposed a YOLO (Redmon et al 2016) based method to detect components in scanned structural diagrams.…”

Section: Application Domainsmentioning

confidence: 99%

“…Furthermore, there can be scientific annotations and the drawings can be edited over time to contain annotations from multiple disciplines. These diagrams are prevalent across multiple industries, including electrical (De et al 2011), oil and gas (Elyan et al 2020a), and architecture (Kim et al 2021a). Manual analysis of these diagrams is time-consuming, prone to human error (Paliwal et al 2021a, b) and requires subject matter experts (Paliwal et al 2021a).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A review of deep learning methods for digitisation of complex documents and engineering diagrams

Jamieson,

Francisco Moreno-García,

Elyan

2024

Artif Intell Rev

View full text Add to dashboard Cite

This paper presents a review of deep learning on engineering drawings and diagrams. These are typically complex diagrams, that contain a large number of different shapes, such as text annotations, symbols, and connectivity information (largely lines). Digitising these diagrams essentially means the automatic recognition of all these shapes. Initial digitisation methods were based on traditional approaches, which proved to be challenging as these methods rely heavily on hand-crafted features and heuristics. In the past five years, however, there has been a significant increase in the number of deep learning-based methods proposed for engineering diagram digitalisation. We present a comprehensive and critical evaluation of existing literature that has used deep learning-based methods to automatically process and analyse engineering drawings. Key aspects of the digitisation process such as symbol recognition, text extraction, and connectivity information detection, are presented and thoroughly discussed. The review is presented in the context of a wide range of applications across different industry sectors, such as Oil and Gas, Architectural, Mechanical sectors, amongst others. The paper also outlines several key challenges, namely the lack of datasets, data annotation, evaluation and class imbalance. Finally, the latest development in digitalising engineering drawings are summarised, conclusions are drawn, and future interesting research directions to accelerate research and development in this area are outlined.

show abstract

“…Compared to these datasets, those used in automatic oor plan analysis studies [22][23][24] contain images with clear pixel intensity distinctions between the foreground and background. Fig.…”

Section: Datasetmentioning

confidence: 99%

Machine Learning-based Binarization Technique of Hand-drawn Floor Plans

Suh

Kim

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Purpose: In this study, we propose a two-step binarization method for hand-drawn architectural floor plans to transform them into usable formats for indoor spatial modeling.Methods: First, a Gaussian mixture modeling was adopted to remove texture-like noise from the background. Second, 24 features were extracted to train the random forest model and the remaining line or spot-like noise was removed from the image. Moreover, the proposed method was applied to a completely different architectural drawing set to evaluate its generalization performance.Results: The experimental results indicated that the proposed method outperformed the other binarization techniques. Moreover, binarization result was outperforming with 0.987 F1-score. Conclusion: The experimental results showed that the overall performance of the proposed method was significantly superior to that of the other binarization methods. Moreover, they indicated that the proposed method is applicable to different types of architectural drawing, thereby proving its generalization.

show abstract

Automatic Extraction of Indoor Spatial Information from Floor Plan Image: A Patch-Based Deep Learning Methodology Application on Large-Scale Complex Buildings

Cited by 11 publications

References 26 publications

Methods and Applications of Space Understanding in Indoor Environment—A Decade Survey

Methods and Applications of Space Understanding in Indoor Environment—A Decade Survey

A review of deep learning methods for digitisation of complex documents and engineering diagrams

Machine Learning-based Binarization Technique of Hand-drawn Floor Plans

Contact Info

Product

Resources

About