Image segmentation is an essential step involved in most computer vision systems used in several areas such as engineering, communications, transportation, business, and social sciences. Besides, several techniques are employed to accomplish the task of segmentation for example (Threshold Based, Region Based, Edge Based, Clustering Based, Neural Network, and Partial Differential Equation). However, thresholding based techniques are the most frequently used techniques for segmentation due to its simplicity, easy to understand and implement. Thresholding in simple words can be defined as the process of dividing an image into multiple homogeneous classes. Mainly, thresholding techniques can be classified into two main categories namely Bi-level Thresholding (BT) and Multi-level Thresholding (MT) based on the number of thresholds selected. BT methods aim to find a single optimal threshold value that separates an image based on each pixel intensity into two regions namely foreground and background. While the MT methods are used to divide an image into two or more classes using two or more thresholds. However, to accomplish the selection task of the optimal thresholds traditional thresholding methods are exhaustive and time-consuming. On the other hand, optimization algorithms proved its efficiency in solving several real-world issues, which make the use of optimization algorithms a promising way of solving thresholding problems. This study presents a literature review of nature inspired optimization algorithms (NIOAs) and its applications in thresholding to optimize the selection of optimal thresholds process based image segmentation domain, also covers a large-scale of the researches published in this area of image segmentation based NIOAs over the last ten years.