Object detection is a vast domain which comprises of various methods of detecting objects efficiently and with good accuracy. Most of the images that we use for detection of objects are RGB images. But what if the images that we are using are captured during adverse weather conditions like foggy, hazy or in bad illumination conditions (during night), our models for object detection might not perform well in such cases. To overcome the problem of bad illumination, images can be captured by an Infrared (IR) camera which is robust under such condition. Thermal images are captured through the Infrared camera. Thermal Images capture the image with the radiation (heat) into visible images, such images are useful in many applications like autonomous driving, military applications, object detection in adverse weather conditions etc. It is essential to make use of thermal images and exploit its use to solve many real-world problems. But due to adverse weather conditions some objects might not be captured properly leading to low accuracy in object detection. To overcome this, we propose a methodology which uses a combination of two methods to enhance thermal images captured in adverse weather conditions which leads to an increase in the accuracy in detection of objects. In this research we also apply and compare the performance of some state-of-the-art object detection models in our use case. Our work primarily uses a thermal dataset called ZUT-FIR-ADAS (Tumas et al. in ZUT-FIR-ADAS, 2020) containing images and its related data captured during adverse weather condition like drizzle, frost, cloudy etc. and applies robust deep learning models like YOLO, Detectron2 etc. The output of this model will help to recognize objects in thermal images efficiently. This research can be used in various fields to enhance accuracy in detection of objects.