Stimulated Raman scattering (SRS) is a powerful imaging technique that has become popular during the last decades for its ability to image species specific in a sample with high accuracy. The purpose of this thesis is twofold. Firstly, to demonstrate 3D spatial control of where in the sample SRS is generated. Secondly, the spatial behavior of the SRS generation is investigated by experiments and simulations. SRS is a nonlinear scattering phenomenon that is produced when a sample is illuminated with two laser beams, called Stokes and pump beams, whose frequency difference corresponds to a molecular vibration caused by inelastic scattering of an incoming photon. The Stokes beam will stimulate the scattering of the pump beam photons, which leads to an intensity gain in the Stokes beam and an intensity loss in the pump beam. Imaging of SRS is usually performed by point scanning a sample in a laser scanning microscope by the two laser beams. Thereafter, the image is constructed pixel by pixel by detecting either the gain or the loss. Our aim is to perform direct field of view SRS imaging. Two experimental setups are presented in this thesis, one for the 3D spatial control of SRS and one for the investigation of the spatial generation of SRS.  The working principle of imaging is the same in both setups. A cylindrical sample volume was illuminated with the Stokes beam and the SRS was generated by focusing the pump beam into this volume. The diameter of the illuminated cylinder was around 10 mm. The two beams were combined before the sample using a dichroic mirror and after the sample the pump beam was removed by a second dichroic mirror.  The Stokes light was then image onto a camera providing a field of view of around 9.4 mm by 7.94 mm. A phase spatial light modulator (SLM) was used to control the shape and position of the pump beam in three dimensions (3D) in the illuminated volume. The results show that the SLM allowed for control of the position and shape of the generated SRS signal. In the second experimental setup the pump beam was focused into the sample by a lens and the spatial generation of the SRS was investigated. A second dichroic mirror blocking the pump beam was inserted into the sample at different interaction lengths to study the resulting SRS signal. Further, the pump intensity was varied to study the effect on the physical width of the SRS signal. The experimental results were compared to computer simulations. The simulations were based on diffraction theory for the beam propagation and the interaction between the light beams and the material was modeled with a phase modulation due to the induced Kerr effect caused by high pump intensity. The results shows that most of the SRS generation takes place close to the focus of the pump beam.

Light can be used in a wide range of measurement methods, from probing the structure of a molecule to quality control of industrial-produced products. This thesis is the result of the initial work connected to a research project regarding the development of interferometric imaging of stimulated Raman scattering (InFeRa). The combination of two different imaging methods, interferometry and stimulated Raman scattering, into one measurement system would provide images containing information about a sample's three-dimensional structure together with chemical specificity. This would allow for studies of a sample's chemical and structural changes due to environmental stimuli to be investigated simultaneously. 

Raman spectroscopy is a label-free, non-contact, and non-destructive optical measurement method based on inelastic scattering of light from a material illuminated by a laser. The Raman scattered light is shown in a Raman spectrum that provides information about the material's molecular structure. This allows for precise identification of samples and investigation of changes in the molecular structure and bond strength. In \textbf{Paper Paper A, Raman spectroscopy was used to investigate bacterial nano cellulose (BNC) produced by bacteria feed with three different carbon sources: glucose, terephthalic acid (TPA), and ethylene glycol (EG). The results were compared to measurements of commercially produced cellulose from plant matter. The investigation showed changes to the glycidic bond responsible for connecting glucose rings with cellulose chains. 

Stimulated Raman scattering (SRS) is a powerful imaging technique that has become popular during the last decades for its ability to image specific species in a sample with high accuracy. When a sample's Raman spectrum is known, SRS can be used to image targeted molecular vibrations in a sample to distinguish between different compounds or to track molecular changes due to environmental stimuli. In SRS, two laser beams, called the Stokes and the pump beams, respectively, are used. SRS can be achieved if the wavelength difference between the two laser beams corresponds to a Raman molecular vibration mode. SRS images are today commonly produced using scanning microscopes, limiting the type of samples that can be imaged.  Implementing an SRS imaging system without relying on high-numerical-aperture microscope objectives and scanning stages would enable the investigation of samples that do not fit under a microscope. If SRS imaging is to be realized in 3D using cameras, understanding the 3D spatial generation of the signal is important. 

It is also of interest to be able to control where the signal is generated in the sample. The working principle of the experiments was to illuminate a cylindrical volume of the sample with the Stokes beam and to use a lens to focus the pump beam into this volume. In Paper B the spatial generation in ethanol was investigated by experiments and simulations. A dichroic mirror was inserted into the ethanol sample to separate the pump, and Stokes beam at different positions, thus stopping the generation of SRS and allowing the resulting signal to be studied. Further, the pump intensity was varied to study the effect on the spatial width of the SRS signal. The experimental results were further compared to computer simulations. The simulations were based on diffraction theory for the beam propagation, and the interaction between the light beams and the material was modeled with a phase modulation due to the induced Kerr effect caused by the high pump intensity. The results show that most of the SRS generation takes place close to the focus of the pump beam. In Conference contribution A, a phase spatial light modulator (SLM) was integrated into the SRS experimental setup to allow for spatial control of the generated SRS signal. It was shown that the shape and position of the pump beam focus and, thereby, the shape of the generated SRS signal could be controlled.

Interferometric imaging can provide information on a sample's 3D structure in various applications. However, an alternative approach was explored due to certain limitations and requirements of SRS imaging.SRS imaging with cameras requires the capturing of two images: one containing the SRS signal and a reference image without the signal. Ideally, these images should be recorded simultaneously, which is not possible with the system used in Paper B and Conference contribution A. Another limitation is the sensitivity of an interferometric imaging system to disturbances. To address these limitations, a robust dual-camera imaging system that provides phase contrast imaging based on digital image correlation of speckles was demonstrated in Paper C. The system utilizes the light deflection in an image's Fourier plane, which has a linear relation to the refraction strength of a phase object and can be calculated from the speckle correlation between the two cameras. The phase object was reconstructed using digital image correlation for a large number of independent speckle patterns. The system was evaluated with experiments and simulations. The system was shown to produce a quantitative measurement of the phase objectives' refraction strength with sensitivity on par with digital holographic interference measurements. However, some errors in the reconstruction were identified to be caused by an out-of-plane movement of the speckle field. 

The image quality of the reconstructed image is internally linked to the standard deviation in the correlation value, and the number of images used in the image reconstruction, which is investigated in Papper D. The standard deviation in the correlation values for different numbers of images used in the reconstruction was investigated for different f-numbers with experiments and simulations. The results show that the f-number of the camera objective controls the imaging system sensitivity but does not directly impact the precision of the measurements. The standard deviation drops off as 1/N^1/2 where is the number of images used in the reconstruction.