Exploiting the early diffusive part of the pulse in object detection in random media

2014 ◽  
Author(s):  
E. H. Bleszynski ◽  
M. K. Bleszynski ◽  
T. Jaroszewicz
Keyword(s):  
Object Detection ◽  
Random Media ◽  
Diffusive Part

Random media with temperature controlled connectivity

1992 ◽  
Vol 2 (5) ◽  
pp. 503-510 ◽  
Author(s):  
F. Carmona ◽  
E. Valot ◽  
L. Servant ◽  
M. Ricci
Keyword(s):  
Random Media ◽  
Temperature Controlled

Automatic accounting of Baikal diatomic algae: approaches and prospects

2019 ◽  
pp. 295-299
Author(s):  
Кonstantin А. Elshin ◽  
Еlena I. Molchanova ◽  
Мarina V. Usoltseva ◽  
Yelena V. Likhoshway
Keyword(s):  
Object Detection ◽  
Loss Function ◽  
Classification Accuracy ◽  
Diatom Species ◽  
Bounding Box ◽  
Synedra Acus ◽  
And Training

Using the TensorFlow Object Detection API, an approach to identifying and registering Baikal diatom species Synedra acus subsp. radians has been tested. As a result, a set of images was formed and training was conducted. It is shown that аfter 15000 training iterations, the total value of the loss function was obtained equal to 0,04. At the same time, the classification accuracy is equal to 95%, and the accuracy of construction of the bounding box is also equal to 95%.

Object Detection in Cluttered Range Images Using Edgel Geometry

2010 ◽  
Vol 130 (9) ◽  
pp. 1572-1580
Author(s):  
Dipankar Das ◽  
Yoshinori Kobayashi ◽  
Yoshinori Kuno
Keyword(s):  
Object Detection ◽  
Range Images

Investigation of particulate systems using optical pathlength spectroscopy

2000 ◽  
Vol 627 ◽  
Author(s):  
Gabriel Popescu ◽  
Aristide Dogariu
Keyword(s):  
Path Length ◽  
Random Media ◽  
Michelson Interferometer ◽  
Industrial Process ◽  
Optical Path Length ◽  
Industrial Applications ◽  
Optical Path ◽  
Interference Signal ◽  
Optical Length ◽  
Reference Mirror

ABSTRACTIn many industrial applications involving granular media, knowledge about the structural transformations suffered during the industrial process is desirable. Optical techniques are noninvasive, fast, and versatile tools for monitoring such transformations. We have recently introduced optical path-length spectroscopy as a new technique for random media investigation. The principle of the method is to use a partially coherent source in a Michelson interferometer, where the fields from a reference mirror and the sample are combined to obtain an interference signal. When the system under investigation is a multiple-scattering medium, by tuning the optical length of the reference arm, the optical path-length probability density of light backscattered from the sample is obtained. This distribution carries information about the structural details of the medium. In the present paper, we apply the technique of optical path-length spectroscopy to investigate inhomogeneous distributions of particulate dielectrics such as ceramics and powders. The experiments are performed on suspensions of systems with different solid loads, as well as on powders and suspensions of particles with different sizes. We show that the methodology is highly sensitive to changes in volume concentration and particle size and, therefore, it can be successfully used for real-time monitoring. In addition, the technique is fiber optic-based and has all the advantages associated with the inherent versatility.

Sign in / Sign up

Export Citation Format

Share Document