Introduction Developing countries often have limited funds and thus diagnostic and treatment modalities should be operated in a cost effective manner while maintaining proper patient care. Gamma camera collimators are acquired at additional cost and the pinhole collimator is often more expensive than the more commonly used Low-energy-high-resolution (LEHR) collimator. The aim of this study was to compare the image quality of a pinhole and parallel-hole collimator using a thyroid phantom. Materials and Methods The thyroid phantom is a perspex block of 10 × 10 cm with thickness of 3 cm. It contained a cavity of an average human thyroid with one hot spot and three cold spots of different sizes and locations. A defect was placed on the phantom at different locations and 53 MBq Tc-99m was injected into the phantom and imaged using the following imaging protocols; (i) clinical protocol as specified by the manufacturer, (ii) equivalent count densities for the pinhole and parallel hole, (iii) equivalent zoom factor for both collimators, (iv) variable detector to phantom distance for the parallel hole collimator. The images were assessed for defect detectability, contrast and sensitivity using human observers. Results All the observers could find the defect for all imaging protocols when the pinhole collimator was used, while in the case of the parallel hole collimator this was only possible when the detector to phantom distance was decreased. The pinhole collimator had superior contrast to that of the parallel hole collimator. As expected the sensitivity of the pinhole collimator was less than that of the parallel hole collimator. Conclusion From this study it is evident that the image quality of the pinhole collimator is superior to the parallel-hole collimator and therefore acquiring a pinhole collimator for thyroid imaging on the gamma camera is justified.