We analyze the poloidal distribution of impurity ions, such as C V, C VI and O VIII, in fast rotating KSTAR plasmas using the impurity emission lines in the 1–7 nm wavelength range obtained from the space-resolved extreme ultra violet spectrometer. This is aided particularly by a new tomographic reconstruction technique with limited viewing angles. The vacuum and extreme ultraviolet spectrometer system at ITER and its prototypes installed at KSTAR are one of the main diagnostic systems to monitor the impurity species in fusion plasmas; however, they have an insufficient field of view to utilize the conventional reconstruction method. The new algorithm developed in this work considers a curvilinear grid of pixels mapped onto the magnetic equilibrium and a weight factor considering toroidal rotational effect on impurity species. Validation tests with poloidally asymmetric images of synthetic plasma demonstrate that this method is highly useful for emission diagnostics that have a narrow field of view. The experimental results of the spatially resolved spectra from KSTAR plasma clearly show that even the light impurities such as carbon and oxygen are significantly affected by the toroidal rotation. Accordingly, the method is expected to be utilized for estimating the poloidal density profile of each charge state of an impurity species, which will play a key role in impurity transport studies.