FPGA architecture for text extraction from images

Vignesh, O.; Mangalam, H.; Gayathri, S.

doi:10.1007/s10586-017-1567-z

Cited by 8 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An NVIDIA GTX TitanX GPU was used for training purposes. In [ 65 ], Vignesh et al applied a morphological closing operation and connected component analysis to detect and extract texts from an image. The authors used a Virtex7 FPGA device to synthesize and evaluate their architecture.…”

Section: Related Workmentioning

confidence: 99%

iDocChip: A Configurable Hardware Accelerator for an End-to-End Historical Document Image Processing

et al. 2021

View full text Add to dashboard Cite

In recent years, there has been an increasing demand to digitize and electronically access historical records. Optical character recognition (OCR) is typically applied to scanned historical archives to transcribe them from document images into machine-readable texts. Many libraries offer special stationary equipment for scanning historical documents. However, to digitize these records without removing them from where they are archived, portable devices that combine scanning and OCR capabilities are required. An existing end-to-end OCR software called anyOCR achieves high recognition accuracy for historical documents. However, it is unsuitable for portable devices, as it exhibits high computational complexity resulting in long runtime and high power consumption. Therefore, we have designed and implemented a configurable hardware-software programmable SoC called iDocChip that makes use of anyOCR techniques to achieve high accuracy. As a low-power and energy-efficient system with real-time capabilities, the iDocChip delivers the required portability. In this paper, we present the hybrid CPU-FPGA architecture of iDocChip along with the optimized software implementations of the anyOCR. We demonstrate our results on multiple platforms with respect to runtime and power consumption. The iDocChip system outperforms the existing anyOCR by 44× while achieving 2201× higher energy efficiency and a 3.8% increase in recognition accuracy.

show abstract

Section: Related Workmentioning

confidence: 99%

iDocChip: A Configurable Hardware Accelerator for an End-to-End Historical Document Image Processing

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The image depicts the tissue's vascular and functional data. The creation of imaging technologies as a consequence of increasing R&D investments and the rapid speed of technology innovations is what is largely fueling the expansion of the worldwide experimental imaging market [4]. By the conclusion of 2018, it is predicted that the worldwide high-resolution image processing market will reach over $1.9 billion, increasing at an 11.37%.…”

Section: Introductionmentioning

confidence: 99%

Reconstructing the Photoacoustic Image with High Quality using the Deep Neural Network Model

Vignesh¹,

Anusuya²,

Senthilkumar³

et al. 2023

Contrast Media & Molecular Imaging

Self Cite

View full text Add to dashboard Cite

The range of diagnostic equipment has been widened and improved by the quick development of biomedical research technologies. The creation of multifunctional instruments that become essential for biomedical operations has been discovered by several research organizations to be made possible by optical imaging, acoustic image analysis, and magnetic resonance imaging. One of the most crucial tools is hyperspectral photoacoustic (PA) imaging, which combines optical and ultrasonic technology. In this study, the reconstruction of the PA pictures employs a new deployment of deep learning methods. This enabled us to train and evaluate our deep-learning approach under several imaging situations in addition to firmly establishing the contextual information. This study presents an optimization approach that blends multispectral optical acoustic imaging with detailed transfer learning-based diagnostic imaging. The particle swarm-convolutional neural network (PS-CNN) technique aims to reconstruct and categorize the presence of cancer using ultrasonic pictures. In image processing, the technique of bilateral filtration (BF) is commonly employed to remove noise. Additionally, the biological images are separated using portable LED Net frameworks. It is also possible to employ a feature extraction technique with the PS optimization methodology. Last but not least, biological images employ a CNN model to assign suitable classification. Using a standard dataset, the PS-CNN technology’s efficacy is confirmed, and testing findings revealed that it performs superior to other methods.

show abstract

Smart Shopping Trolley based on IoT and AI for the Visually Impaired

Selvan,

Stella,

et al. 2024

2024 International Conference on Cognitive Robotics and Intelligent Systems (ICC - ROBINS)

View full text Add to dashboard Cite

FPGA architecture for text extraction from images

Cited by 8 publications

References 14 publications

iDocChip: A Configurable Hardware Accelerator for an End-to-End Historical Document Image Processing

iDocChip: A Configurable Hardware Accelerator for an End-to-End Historical Document Image Processing

Reconstructing the Photoacoustic Image with High Quality using the Deep Neural Network Model

Smart Shopping Trolley based on IoT and AI for the Visually Impaired

Contact Info

Product

Resources

About