Approach for Document Detection by Contours and Contrasts

Tropin, Daniil V.; Ilyuhin, Sergey A.; Nikolaev, Dmitry P.; Arlazarov, Vladimir V.

doi:10.1109/icpr48806.2021.9413271

Cited by 14 publications

(23 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This score is based on the integral characteristics of the contours along the sides {b} of the quadrilateral q: the intensity of the edges inside w and outside w of the borders [8], and the shares of non-zero pixels c on the edge map along borders b [5]. In [10], Tropin et al suggested the following contour score:…”

Section: Ranking Quadrilaterals By Contours and Contrastsmentioning

confidence: 99%

“…To improve the performance on complex backgrounds (Fig. 3a), the authors of [10] employed the contrast estimation of the quadrilateral candidates. It is based on the  2 distance between the RGB pixel histograms of the inner and outer areas of the quadrilateral.…”

Section: Ranking Quadrilaterals By Contours and Contrastsmentioning

confidence: 99%

“…The systems proposed in [7,8] do not have reference performance evaluations on open datasets. The methods [10,11] were evaluated on the open dataset SmartDoc [23]. And in spite of the high-quality performance, they do not achieve state-of-the-art scores.…”

Section: Introductionmentioning

confidence: 99%

“…And in spite of the high-quality performance, they do not achieve state-of-the-art scores. The performance of the algorithms proposed in [9] as well as in [10] was evaluated on the open MIDV-500 dataset [24].…”

Section: Introductionmentioning

confidence: 99%

“…In this paper, we propose the on-device document localization system, which also takes a Hough-based approach. We will use the modifications proposed in [9,10] to make the system more reliable in terms of FP and BV problems. The proposed method allowed for the second best result for SmartDoc dataset in terms of precision, surpassed only by a U-net like neural network.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Advanced Hough-based method for on-device document localization

et al. 2021

View full text Add to dashboard Cite

The demand for on-device document recognition systems increases in conjunction with the emergence of more strict privacy and security requirements. In such systems, there is no data transfer from the end device to a third-party information processing servers. The response time is vital to the user experience of on-device document recognition. Combined with the unavailability of discrete GPUs, powerful CPUs, or a large RAM capacity on consumer-grade end devices such as smartphones, the time limitations put significant constraints on the computational complexity of the applied algorithms for on-device execution. In this work, we consider document location in an image without prior knowledge of the docu-ment content or its internal structure. In accordance with the published works, at least 5 systems offer solutions for on-device document location. All these systems use a location method which can be considered Hough-based. The precision of such systems seems to be lower than that of the state-of-the-art solutions which were not designed to account for the limited computational resources. We propose an advanced Hough-based method. In contrast with other approaches, it accounts for the geometric invariants of the central projection model and combines both edge and color features for document boundary detection. The proposed method allowed for the second best result for SmartDoc dataset in terms of precision, surpassed by U-net like neural network. When evaluated on a more challenging MIDV-500 dataset, the proposed algorithm guaranteed the best precision compared to published methods. Our method retained the applicability to on-device computations.

show abstract

Section: Ranking Quadrilaterals By Contours and Contrastsmentioning

confidence: 99%

Section: Ranking Quadrilaterals By Contours and Contrastsmentioning

confidence: 99%