• Madhvi M Buha
• Prof. Daxa Vekariya
• Prof. Shradhdha Bhalodiya
• Dr. Vipul Vekariya

Images captured beneath outside scenes sometimes suffer from low distinction and restricted visibility thanks to suspended region particles, that directly affects the standard of photos. Despite varied image dehazing ways are projected, effective hazy image restoration remains a difficult downside. Existing learning-based ways sometimes predict the medium transmission by Convolutional Neural Networks (CNNs), however ignore the key international region light-weight. completely different from previous learning-based ways, we have a tendency to propose a versatile cascaded CNN for single hazy image restoration, that considers the medium transmission and international region light-weight collectively by 2 task-driven subnet works. Specifically, the medium transmission estimation subnetwork is galvanized by the densely connected CNN whereas the worldwide region light-weight estimation subnetwork may be a light-weight CNN. Besides, these 2 subnetworks square measure cascaded by sharing the common options. Finally, with the calculable model parameters, the haze-free image is obtained by the region scattering model inversion, that achieves a lot of correct and effective restoration performance. Qualitatively and quantitatively experimental results on the artificial and real-world hazy pictures demonstrate that the projected methodology effectively removes haze from such pictures, and outperforms many progressive dehazing ways

Image haze removal, image enhancement, DCP,CNN,RCNN

"Augmented Approach On Single Image Dehazing Using R-CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.530-537, April-2019, Available :http://www.jetir.org/papers/JETIR1904693.pdf

"Augmented Approach On Single Image Dehazing Using R-CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp530-537, April-2019, Available at : http://www.jetir.org/papers/JETIR1904693.pdf