Inversion of the radiative transfer equation using artificial neural network

literature review provides what is the state of the art currently in addressing this problem what has the people done or what different kind of methods they used and then what they specifically done with spectroscopy and optical properties and what the problems they faced also what types of networks and methods they used for example paper X used this technique while another paper used some other techniques and what were are the technical issues. we need to talk about other people’s work and what they have been done we definitely need to give a reference because it is not our work, if you are making a statement why this cannot be use there must be an evidence someone must has said it in some other paper based on some research which we need to mention.

The literature review should answer all the question bellow:

Has neural networks been used widely in spectroscopy? If no why only few people are using neural network in spectroscopy? If yes why? Are they using some other methods other than neural network more commonly in spectroscopy? What other methods were people using apart from neural networks in spectroscopy? What are the most commonly used methods? Did those work or didn’t work and why? Is neural network having better results or worst results than those methods and why?

Same review needs to be done for optical properties. And same questions should be answered for optical properties

Has neural networks been used widely in optical properties? If no why only few people are using neural network in optical properties? If yes why? Are they using some other methods other than neural network more commonly in optical properties? What other methods were people using apart from neural networks in optical properties? What are the most commonly used methods? Did those work or didn’t work and why? Is neural network having better results or worst results than those methods and why?

Reminding about thesis topic:

 

Inversion of the radiative transfer equation using machine learning techniques (Electrical Engineering)

Radiative transfer of energy (thermal and electromagnetic radiation) propagating through a medium containing particulate matter is important in many areas of science and engineering. In atmospheric sciences, the impact of smog and environmental pollutants are investigated by analysing solar radiation propagating through the atmosphere. Radiative heat transfer plays an important role in the optimal design of furnaces and radiators. In medical diagnostics using optical methods, the propagation of electromagnetic radiation is used to characterise tissue and blood for example, to screen for cancer and to monitor wounds. More recently, this approach is also being considered for characterising chemical suspensions and powder mixtures. In all these applications, radiative transfer theory (RTT) is used to describe the propagation of radiation energy.

The radiative transfer equation (RTE) arising out of this theory is computationally intensive particularly if the intention is to extract the radiative transfer coefficients which contain information regarding the characteristics of the medium through which the radiation has travelled. In this study, the RTE solution will be obtained

by using machine learning techniques such as Neural Networks, Support Vector Machines etc. The performance of these methods will be compared for solving the RTE under different conditions. This work will be carried out using Matlab and the machine learningtoolbox available in Matlab. Data required for this work will be generated using a simulation software in Matlab that is available in-house.

 

In my thesis what I’ll be working on is that we have the input obtained from a spectroscopic measurement which is done by sending light thru a sample and we collecting the lights gone thru the sample example transmitted light, reflected light, different angles whatever. After collecting measurement we want to know how many particles per unit volume has that and more importantly we need to know the size of the particle.

In our case we need to build a neural network which by giving the measurements taken by the spectroscopy the network will tell us for each particular wavelength what is the absorption coefficient and what is the scattering coefficient.

 

For a custom paper on the above or a related assignment, place your order now!

Use  “20” as the discount code to redeem your offer!

What We Offer:

• Affordable Rates – (15 – 25% Discount on all orders above $50 )
• 100% Free from Plagiarism
• Masters & Ph.D. Level Writers
• Money Back Guarantee