This paper addresses the problem of automatic detection and identification of celebrity cartoon faces and corpus creation. However, currently available face detection systems are mainly designed to detect human faces. We developed a celebrity car-toon face detector (CCFC) that specially detects celebrity cartoon faces featuring famous individuals, which is customised by a well-known object detector, Yolo-NAS, a highly efficient and lightweight deep architecture. Initially, we train our detector by employing a manually compiled dataset. Also, a suitable matching technique is utilised to create a celebrity cartoon face image database. The performance of the proposed system is evaluated with three different matching thresholds, and 90% accuracy is achieved in database creation. The novelty of the work lies in the complete setup of the process of creating a deployable database since this work is the first of its kind to automate the task of creating a corpus of images.