The increased range of manufacturable freeform surfaces offered by the new fabrication techniques is giving opportunities to incorporate them in the optical systems. However, the success of these fabrication techniques depends on the capabilities of metrology procedures and a feedback mechanism to CNC machines for optimizing the manufacturing process. Therefore, a precise and in-situ metrology technique for freeform optics is in demand. Though all the techniques available for aspheres have been extended for the freeform surfaces by the researchers, but none of the techniques has yet been incorporated into the manufacturing machine for in-situ measurement. The most obvious reason is the complexity involved in the optical setups to be integrated in the manufacturing platforms. The Shack-Hartmann sensor offers the potential to be incorporated into the machine environment due to its vibration insensitivity, compactness and 3D shape measurement capability from slope data. In the present work, a measurement scheme is reported in which a scanning Shack-Hartmann Sensor has been employed and used as a metrology tool for measurement of freeform surface in reflection mode. Simulation studies are conducted for analyzing the stitching accuracy in presence of various misalignment errors. The proposed scheme is experimentally verified on a freeform surface of cubic phase profile.
Visible - Near-infrared spectroscopy (Vis-NIRS) is now commonly used to measure different physical and chemical parameters of soils, including carbon content. However, prediction model accuracy is insufficient for Vis-NIRS to replace routine laboratory analysis. One of the biggest issues this technique is facing up to is light scattering due to soil particles. It causes departure in the assumed linear relationship between the Absorbance spectrum and the concentration of the chemicals of interest as stated by Beer-Lambert's Law, which underpins the calibration models. Therefore it becomes essential to improve the metrological quality of the measured signal in order to optimize calibration as light/matter interactions are at the basis of the resulting linear modeling. Optics can help to mitigate scattering effect on the signal. We put forward a new optical setup coupling linearly polarized light with a Vis-NIR spectrometer to free the measured spectra from multi-scattering effect. The corrected measured spectrum was then used to compute an Absorbance spectrum of the sample, using Dahm's Equation in the frame of the Representative Layer Theory. This method has been previously tested and validated on liquid (milk+ dye) and powdered (sand + dye) samples showing scattering (and absorbing) properties. The obtained Absorbance was a very good approximation of the Beer-Lambert's law absorbance. Here, we tested the method on a set of 54 soil samples to predict Soil Organic Carbon content. In order to assess the signal quality improvement by this method, we built and compared calibration models using Partial Least Square (PLS) algorithm. The prediction model built from new Absorbance spectrum outperformed the model built with the classical Absorbance traditionally obtained with Vis-NIR diffuse reflectance. This study is a good illustration of the high influence of signal quality on prediction model's performances. 2b1af7f3a8