A combined 3D and hyperspectral method for surface imaging of wounds

Paluchowski, Lukasz A.; Denstedt, Martin; Røren, Thomas; Pukstad, Brita; Randeberg, Lise Lyngsnes

doi:10.1117/12.2001248

Cited by 4 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Capítulo 5: Aportaciones de la combinación de técnicas 3D e hiperespectrales Howell, & Schneider, 2013). En estos estudios se realiza la fusión de datos usando láseres LIDAR de tiempo de vuelo con sensores hiperespectrales, sin embargo, esta fusión se puede también realizar con otro tipo de sensores tridimensionales como demuestra la combinación de una cámara estéreo con una cámara hiperespectral para el estudio de heridas superficiales (Paluchowski, Denstedt, Røren, Pukstad, & Randeberg, 2013).…”

Section: -Introducciónunclassified

Desarrollo de técnicas de visión hiperespectral y tridimensional para el sector agroalimentario.

Martínez¹

View full text Add to dashboard Cite

com els ulls on es pot estimar millor el frescor. Els resultats de R2 CV van ser de 0.844 al correlacionar els espectres obtinguts dels ulls amb el temps mitjançant un PLSDA.El desenvolupament d'aquestes tècniques tindrà un gran impacte en la indústria agroalimentària en un futur pròxim, ja que suposa una clara innovació tecnològica respecte a realitzar anàlisis físic-químics destructius en un subconjunt de les mostres. Aquestes tècniques permeten realitzar el control de qualitat i de seguretat de totes les mostres de forma no destructiva millorant per tant la qualitat, rapidesa, seguretat, fiabilitat i cost dels diferents processos de la indústria alimentària.Página ix de xix Abstract AbstractEnhancements of the productive processes in the food industry, as any other industry, are a key factor to keep the competitiveness and increase profits. In order to achieve this objective it is needed to use new methods that improve the quality and efficiency of these processes.Moreover, nowadays there is a farther distance between the place of production of the current foods and the place where they are eaten. This requires quality systems that can inspect the 100% of the food samples in a cheap and non-destructive way. Hyperspectral and 3D techniques are proposed in this thesis as a solution.A review of the current state of art has been done for the different techniques to obtain three dimensions´ information as well as their uses in the food industry. Structured light (SL), stereo vision and time of flight (TOF) have been chosen as the best suitable. A comparison between SL and TOF for the in-line measurement of three animal foods and three vegetable foods has been done. The conclusion of this study is that both techniques are suitable to use having a mean R 2 CV of 0.85 for TOF and 0.95 for SL for the volume estimation of the samples. SL techniques have been studied deeper solving the segmentation problem of detecting roots on potatoes. This is a difficult problem to solve using classic computer vision techniques due to the similar colors between the roots and the potatoes. This problem was solved doing an Adaboot model that classifies the 3D points of the cloud into roots or surface points using a 3D features vector for each point. The results achieved values of 94% in accuracy. Another problem solved was the assessment of grape cluster yield components based on 3D descriptors using stereo vision. Compactness is especially difficult to assess due its subjectivity. Currently, this quality component is evaluated by a group of experts following the method described in OIV volume 204. A semi-automatic method was developed to solve this problem using new 3D descriptors and a SVM model. A prediction value of R 2 =0.8 was achieved for hundred bunches of ten different varieties. Página x de xix AbstractWith regard of hyperspectral techniques, a new methodology has been developed in order to get results from hyperspectral images. This methodology has been applied for solving three different problems. These problems were freshne...

show abstract

Section: -Introducciónunclassified

Desarrollo de técnicas de visión hiperespectral y tridimensional para el sector agroalimentario.

Martínez¹

View full text Add to dashboard Cite

show abstract

Diagnostic and Prognostic Utility of Non-Invasive Multimodal Imaging in Chronic Wound Monitoring: a Systematic Review

Mukherjee

Tewary

Routray³

2017

J Med Syst

View full text Add to dashboard Cite

Monitoring chronic wound [CW] healing is a challenging issue for clinicians across the world. Moreover, the health and cost burden of CW are escalating at a disturbing rate due to a global rise in population of elderly and diabetic cases. The conventional approach includes visual contour, sketches, or more rarely tracings. However, such conventional techniques bring forth infection, pain, allergies. Furthermore, these methods are subjective as well as time-consuming. As such, nowadays, non-touching and non-invasive CW monitoring system based on imaging techniques are gaining importance. They not only reduce patients' discomfort but also provide rapid wound diagnosis and prognosis. This review provides a survey of different types of CW characteristics, their healing mechanism and the multimodal non-invasive imaging methods that have been used for their diagnosis and prognosis. Current clinical practices as well as personal health systems [m-health and e-health] for CW monitoring have been discussed.

show abstract