FEMT: a computational approach for fog elimination using multiple thresholds
- PDF / 1,368,869 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 14 Downloads / 200 Views
FEMT: a computational approach for fog elimination using multiple thresholds Mamta Mittal 1 & Munish Kumar 2 & Amit Verma 3 & Iqbaldeep Kaur 3 & Bhavneet Kaur 4 & Meenakshi Sharma 4 & Lalit Mohan Goyal 5 Received: 22 October 2019 / Revised: 17 July 2020 / Accepted: 18 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Refining visibility through haze removal from image becomes an inevitable chore and essential to recognize and track vehicles, traffic signal, and signs clearly under road safety. That can face a recurrent degradation under destitute climatic circumstances for instance fog, rain, cloud, and smog. To diminish this constraint, various methods were designed and implemented, but most were not capable of obtaining the improved quantitative outcomes. Therefore, a new algorithm Fog Elimination using Multiple Thresholds (FEMT) for single image haze eviction that meritoriously obtains the significant results on both gray and colored over real and synthetic images using multiple thresholds is proposed in this paper. The proposed method targets on the light regions by reducing the brightness and increasing the contrast of image at different levels. Finally, by grouping all the obtained resultant images leads to the generation of the resultant defogged image. The qualitative and quantitative analysis is carried out for an assessment of digitalized de-hazed images acquired from the proposed algorithm and compared to the prior techniques. Simulated fallouts entitle high resemblance to the corresponding ground truth, reduction in computation time consumption to 88% and error of 98%. The proposed approach can be applied in the field of robotics, human activity monitoring, smart systems, and digital investigation on the hazy images. Keywords Fog Removal . Multiple Thresholds . Image Enhancement . Visibility Restoration
1 Introduction Poor weather sources from the major atmospheric components (such as fog/haze, rain, cloud and smog) turns a diminution in visibility and color distortion in the scenes [11]. Degradation in scene visibility is primary cause of accidents all over the globe. When light approaching
* Munish Kumar [email protected] Extended author information available on the last page of the article
Multimedia Tools and Applications
towards the observer’s vision or camera, it acquires attenuation because of scattering of light through the dews (Fig. 1). According to a study visibility in 300–1000 m range is reviewed as low fog and below 300 m non-visibility is reviewed as dense fog [11]. Fog eliminations from image turn to a thoughtful concern since the partial analysis indications about scene understanding is present [23]. On the basis of study made we identify blurriness in a scene because decrement in visibility leads to reduction in high frequency components. It directly affects image analysis quality because of multiple fog conditions. Fattal [7] performed a fog elimination process over a color image for refinement by projecting Independent Component Analys
Data Loading...
