Image processing using Pearson's correlation coefficient: Applications on autonomous robotics

Neto, A. M. J. C.; Victorino, Alessandro Corrêa; Fantoni, Isabelle; Zampieri, Douglas Eduardo; Ferreira, Janito Vaqueiro; Lima, Danilo Alves de

doi:10.1109/robotica.2013.6623521

Cited by 40 publications

(14 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Another variant of \IDi®Net" was that it utilized negative Pearson correlation coe±cient (NPCC) as the loss function to replace the conventional MAE strategy because NPCC performed better for spatially sparse objects and strong scattering conditions. 45 Experiments with one of two di®erent di®users and training inputs, which were extracted from one of three databases (Faces-LFW, 46 ImageNet, 47 and MNIST 43 ) for each time, were carried out to validate the proposed method. The training size was 10,000 each time.…”

Section: Deep Learning Methods For Optical Tomographymentioning

confidence: 99%

Brief review on learning-based methods for optical tomography

Zhang

2019

J. Innov. Opt. Health Sci.

View full text Add to dashboard Cite

Learning-based methods have been proved to perform well in a variety of areas in the biomedical field, such as biomedical image segmentation, and histopathological image analysis. Deep learning, as the most recently presented approach of learning-based methods, has attracted more and more attention. For instance, massive researches of deep learning methods for image reconstructions of computed tomography (CT) and magnetic resonance imaging (MRI) have been reported, indicating the great potential of deep learning for inverse problems. Optical technology-related medical imaging modalities including diffuse optical tomography (DOT), fluorescence molecular tomography (FMT), bioluminescence tomography (BLT), and photoacoustic tomography (PAT) are also dramatically innovated by introducing learning-based methods, in particular deep learning methods, to obtain better reconstruction results. This review depicts the latest researches on learning-based optical tomography of DOT, FMT, BLT, and PAT. According to the most recent studies, learning-based methods applied in the field of optical tomography are categorized as kernel-based methods and deep learning methods. In this review, the former are regarded as a sort of conventional learning-based methods and the latter are subdivided into model-based methods, post-processing methods, and end-to-end methods. Algorithm as well as data acquisition strategy are discussed in this review. The evaluations of these methods are summarized to illustrate the performance of deep learning-based reconstruction.

show abstract

Section: Deep Learning Methods For Optical Tomographymentioning

confidence: 99%

Brief review on learning-based methods for optical tomography

Zhang

2019

J. Innov. Opt. Health Sci.

View full text Add to dashboard Cite

show abstract

“…The DIC value measurement follows Eq. 1 which is Pearson's Correlation Coefficient where fi and gi are the intensity of the i th pixel in the 1 st image and 2 nd image respectively, fmean and gmean are the mean intensity of the 1 st image and 2 nd image consecutively [11].…”

Section: Methodsmentioning

confidence: 99%

Measurement of Thermal Expansion Coefficient on Electric Cable Using X-Ray Digital Microradiography

Affriyenni¹,

Suparta²,

Swalaganata³

2018

EECSI

View full text Add to dashboard Cite

Electric cable is a medium to conduct electrical energy. Expansion and contraction caused by thermal changes may result in an aging effect on the cable. This paper presents the way to observe the expansion in electrical cable due to thermal changes using the x-ray microradiography. The observed electric cables were NYA, NYAF, and NYM, each with cross-sectional areas of 1.5 mm 2 and 2.5 mm 2. The temperature was monitored using a DS18B20 sensor compiled into a microcontroller. In order to process and analyze the cables images, an ImageJ software was used. The image differences were compared based on the value of the digital image correlation. The physical analysis was carried out based on Adrian's FWHM and calculated using the regression method. The accurate structural dimension measurement using x-ray digital microradiography is about 50 μm/pixel. The average relative error measured was less than 3%.

show abstract

“…Pearson's Correlation (PC) [26]: It gives the measure of how likely it is possible to infer a map from another and it is applied only for occupied space. According to [21] this metric suffers from two drawbacks, one because it requires similar occupied pixels between maps, and the other is that it is perturbed by the outliers (e.g, [27]…”

Section: Benchmarking Metricsmentioning

confidence: 99%

Vector Maps: A Lightweight and Accurate Map Format for Multi-robot Systems

Baizid

Lozenguez

Fabresse

et al. 2016

Intelligent Robotics and Applications

View full text Add to dashboard Cite

SLAM algorithms produce accurate maps that allow localization with typically centimetric precision. However, such a map is materialized as a large Occupancy Grid. Beside the high memory footprint, Occupancy Grid Maps lead to high CPU consumption for path planning. The situation is even worse in the context of multi-robot exploration. Indeed, to achieve coordination, robots have to share their local maps and merge ones provided by their teammates. These drawbacks of Occupancy Grid Maps can be mitigated by the use of topological maps. However, topological maps do not allow accurate obstacle delimitations needed for autonomous robots exploration. So, robots still have to handle with Occupancy Grid Maps. We argue that Vector-based Maps which materialize obstacles using collections of vectors is a more interesting alternative. Vector Maps both provide accurate metric information likewise Occupancy Grid Maps, and represent data as a graph that can be processed for path planning and maps merging as efficiently as with topological maps. Conclusions are backed by several metrics computed with several terrains that differ in size, form factor, and obstacle density. This work was supported by Région Nord Pas-de-Calais as part of the SUCRé project (Human-robots cooperation in hostile environment).

show abstract

Image processing using Pearson's correlation coefficient: Applications on autonomous robotics

Cited by 40 publications

References 27 publications

Brief review on learning-based methods for optical tomography

Brief review on learning-based methods for optical tomography

Measurement of Thermal Expansion Coefficient on Electric Cable Using X-Ray Digital Microradiography

Vector Maps: A Lightweight and Accurate Map Format for Multi-robot Systems

Contact Info

Product

Resources

About