Dynamic Scene Multi-Exposure Image Fusion

ABSTRACT

Exposure fusion is performed on multi-exposure image sequence to obtain the well-exposed image, which contains the details corresponding to the entire dynamic range. However, while capturing multi-exposure image sequence new objects can appear/disappear from the image sequence, thus resulting in ghost effect in the fused image. Elimination of such ghost effect at less computational cost is an important consideration to generate HDR images. In this paper, a simple yet efficient method for multi-exposure image fusion of dynamic scene without ghost effect is proposed. The proposed method is divided into four steps, viz., remapping of images (pre-processing), weight map generation, exposure fusion and color correction with contrast enhancement (post-processing). The experimental result shows that the resultant images produced with the help of the proposed method are free from ghost effect and have good color and contrast. Also, the important thing is that the computational complexity to reach the desired result is very less.

KEYWORDS:

• Exposure fusion
• Ghost effect
• High dynamic range image
• Remapping
• Sigmoid function
• Weight map

ACKNOWLEDGEMENT

We would like to witness our deep and honest gratefulness to Prof. Ashish Vanmali from Department of Electronics and Telecommunication Engineering at Vidyavardhini’s College of Engineering and Technology, University of Mumbai, for his suggestion, help, assistance and vital involvement, which made him an important part of this project. His perceptive, cheering and delicate supervision have provided a good basis for the present project. His involvement with his uniqueness has helped us in gaining knowledge and learning new things that were initially beyond our imagination.

Additional information

Notes on contributors

Uzmanaz A. Shaikh

Uzmanaz A Shaikh is a student in Vidyavardhini’s College of Engineering & Technology, Vasai (W), University of Mumbai, Mumbai, India. Her area of interest is image processing. She is a student member of IETE.

Vivek J. Vishwakarma

Vivek J Vishwakarma is a student in Vidyavardhini’s College of Engineering & Technology, Vasai (W), University of Mumbai, Mumbai, India. His area of interest is image processing. He is a student member of IETE. Email: [email protected].

Shubham S. Mahale

Shubham S Mahale is a student in Vidyavardhini’s College of Engineering & Technology, Vasai (W), University of Mumbai, Mumbai, India. His area of interest is image processing. Email: [email protected].