Iodine value is an important index for evaluating quality of the fatty acid. In order to research the feasibility of measuring unsaturation degree of fatty acid of intact fat sample by using the Raman spectroscopy and chemometrics method, we use pork′s subcutaneous adipose tissue as research object to obtain the Raman spectra in the wavelength range of 300-2000 cm-1 with utilization of self-constructed Raman spectroscopy detection system. Both of preprocessing methods and iodine value related variables are optimized. Using the iodine value obtained by traditional Wijs titration as reference, we build the calibration model of iodine value in fat sample by partial least square algorithm. The experimental results show that the optimal preprocessing method is the 8th order polynomial fitting method. The optimal combination band, 1106-1470 cm-1, is obtained by interval partial least square method. Based on the optimized pretreatment method and selected variables, the iodine value prediction model is developed, the prediction set coefficient (Rp) and root mean square error of prediction (RMSEP) of the model are 0.9463 and 2.5391×10-2, respectively. The study results indicate that optimized variable can be used in rapid and nondestructive measurement of iodine value of intact pork fat, which provides theoretical basis for on-line, rapid, and nondestructive measurement of unsaturation degree of fatty acid in pig carcasses based on Raman spectroscopy.