EDGE AND LAPLACE BASED RESTORATION OF HAZY IMAGES

www.ijrar.om INTERNATIONAL JOURNAL OF RESEARCH IN COMPUTER APPLICATIONS AND ROBOTICS ISSN 2320-7345 EDGE AND LAPLACE BASED RESTORATION OF HAZY IMAGES 1 Priya Singh Patel, 2 Prof. Manisha Rathore Abstrat: - As the digital world is inreasing day by day so numbers of digital image proessing issues are overed by different researhers. Out of those this work fouses on Fog removal whih is also known as visibility restoration refers to different methods that aim to redue or remove the degradation that have ourred while the digital image was being obtained. The degradation may be due to various fators like relative objet-amera motion, blur due to amera miss-fous, relative atmospheri turbulene and others. This paper has utilized the Laplae base distortion detetion with edge information preserving. Combination of both these tehniques helps in identifying the atual olor values present in the original image sene. Experiment is done on many images of different environment or ategory. Results shows that proposed work is better as ompare to previous work in [8]. Index Terms Digital Image Proessing, Haze, Information Extration, Fog removal. Visibility restoration 1. INTRODUCTION Most outdoor visual systems suh as video surveillane, target traking, remote sensing and navigation ontrol and vehile autonomous driving and others are highly vulnerable to harsh environment, espeially beause of fog and haze. Images of outdoor sene an be signifiantly degraded due to the bad weather ondition suh as fog and haze. Thus it lead to the atmospheri sattering of tiny water droplets and atmospheri aerosol on the sene point, ausing image fuzzy, bad visibility, and seriously affeting the performane of an outdoor system. This happens beause of the presene of numerous atmospheri partiles whih absorbs and satters light. Suh degraded images lose all its ontrast and beome dim espeially in the distant regions and get blurred with their surroundings area. In order to make the system robust and reliable in bad weather onditions, it is neessary to dehaze that degraded image. In long distane photography or foggy senes, this proess has a substantial effet on the image in whih ontrasts are redued and surfae olors beome faint. Suh degraded images, photographs lak visual vividness and appeal and moreover, they offer a bad visibility of the ontents of the sene. This may also be the ase for satellite imaging whih is used for various purposes like web mapping, land-use planning and environmental studies et. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 62

www.ijrar.om In this proess light whih should have to travel in straight lines is sattered and replaed by the previously sattered light alled the airlight. The main objetive here is to enhane the images whih are taken under poor visibility and even restore the lear-day visibility of that sene. There are many irumstanes for whih aurate haze removal algorithms are needed. The major goal of the haze removal algorithms is to enhane and reover the detail of the sene from the haze image. Visibility restoration [1] refers to different methods that aim to redue or remove the degradation that have ourred while the digital image was being obtained. The degradation may be due to various fators like relative objetamera motion, blur due to amera misfous, relative atmospheri turbulene and others. In this we will be disussing about the degradations due to bad weather suh as fog, haze, rain and snow in an image. The image quality of outdoor sreen in the fog and haze weather ondition is usually degraded by the sattering of a light before reahing the amera due to these large quantities of suspended partiles (e.g. fog, haze, smoke, impurities) in the atmosphere. This phenomenon affets the normal work of automati monitoring system, outdoor reognition system and intelligent transportation system. Sattering is aused by two fundamental phenomena suh as attenuation and air light. By the usage of effetive haze removal of image we an improve the stability and robustness of the visual system. Haze removal is a tough task beause fog depends on the unknown sene depth information. Fog effet is the funtion of distane between amera and objet. Hene removal of fog requires the estimation of air light map or depth map. The urrent haze removal method an be divided into two ategories: image enhanement and image restoration. Image enhanement does not inlude the reasons of fog degrading image quality. This method an improve the ontrast of haze image but loses some of the information regarding image. Image restoration firstly studies the physial proess of image imaging in foggy weather. After observing that degradation model of fog image will be established. At last, the degradation proess is inverted to generate the fog free image without the degradation. So, the quality of degraded image ould be improved. 2. Related Work [1] introdued an experiene fusion method for various images by way of moving objets. The proposed method onsist a ghost removal algorithm in a low dynami series domain and a exposure fusion algorithm. The proposed ghost removal algorithm inludes a bidiretional normalization-based method for the finding of non-reliable pixels and a two-round hybrid method for the orretion of non-onstant pixels. A exposure fusion algorithm onsist a ontent adaptive bilateral filter, that extrats superior details from all the orreted images onurrently in asent domain. The final image is synthesized by seletively adding the extrated fine details to an in-between image that is generated by fusing all the orreted images via an existing multi-level algorithm. In [2] desribed a novel and effiient single image enhanement algorithm for haze image. As they monitor that, the ontrast and intensity of haze image After using dark hannel prior approah will neessarily tend to be lower than those of the real sene, they used histogram requirement to make an enhanement on image after dark hannel prior approah. They made a large number of experiments and find that, if dealing with a haze image with large Bakground area and low ontrast, dark hannel prior result will beome dark, also a general haze image after dark hannel ours different degree of anamorphous. They introdued an adaptive algorithm to repair the different kinds of an amorphose on the hazy image after dark hannel prior. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 63

www.ijrar.om In [3] projeted a fast yet tough tehnique to enhane the visibility of video frames using the dark hannel prior united with fuzzy logi-based tehnique. The dark hannel prior is a arithmetial uniformity of outdoor haze-free images based on the examination that most loal pathes in the haze-free images have pixels whih are dark in at least one olor hannel, where the fuzzy logi-based tehnique is used to plan an input spae to an output spae using a olletion of fuzzy membership funtions and poliy to deide deliately in ase of doubts. The ombination of the dark hannel and the fuzzy logi-based tehnique will make high quality haze-free images in real-time. In [5] has disussed that the within the last deades, improving the quality of an underwater image has one problem that is poor visibility of the image whih is aroused by physial properties of the water medium. In [6] proposed pereptual models that an be able to foreast the value of distorted images with as little prior information of the images or their deformation as possible. The new IQA model, whih is known as Natural Image Quality Evaluator is based on the prodution of a quality aware olletion of statistial features based on a simple and suessful spae area natural sene statisti model. In [7] proposed novel widespread guided image filtering method with the suggestion image generated by signal subspae projetion tehnique. It aepts ompliated parallel study through Monte Carlo imitation to hoose the dimensions of signal subspae in the path-based noisy images. The noise free image is reonstruted from the noisy image expeted onto the signifiant images by omponent analysis. Test images are utilized to deide the relationship between the most favorable parameter value and noise divergene that maximizes the output peak signal-to-noise ratio. In [8] has presented a new method alled mixture CLAHE olor models that speifially developed for underwater image enhanement. The proess performs CLAHE method on RGB and HSV olor models. The projeted tehnique has onsiderably enhaned the visual superiority of underwater digital images by enhaning illuminate, as well as dropping noise and artifats. 3. Proposed Methodology This paper fous on the digital hazy image restoration. Here image store the edge region of the image then apply Laplae distribution for pixel value restoration. Here whole work is explained in fig. 2. 3.1 Pre-Proessing Here as the image is the olletion of pixels where eah pixel is representing a number that is refleting a number over there now for eah number depend on the format it has its range. So read a image means making a matrix of the same dimension of the image then fill the matrix orrespond to the pixel value of the image at the ell in the matrix. 3.2 Edge Detetion In order to find the edges in the image onvert it into gray format then apply the anny algorithm. This is the method to onvert an gray sale image into binary image. For this analysis of eah pixel is done. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 64

www.ijrar.om Image Dataset Pre-Proessing Canny Algorithm 16X16 Blok De- Fusion Laplae Distribution Adjustable Color Parameters Haze Thikness Estimation Combine Blok Edge Combine Fig. 3. Blok diagram of proposed Restoration Image Work. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 65

www.ijrar.om Edge feature: In ase of edge feature anny algorithm is applied Smooth the Image with Gaussian Filter. Compute the Gradient Magnitude and Orientation using finite-differene approximations for the partial derivatives. Apply non-maxima suppression to the gradient magnitude. Use the double thresholding algorithm to detet and link edges. 16X16 Blok: As work is done on olor image so embedding is done on the red matrix of the image, so whole operation of embedding is done this red matrix. Whole red, green, blue matrix is divided into 16X16 bloks for restoration of image. 3.3 Laplae Distribution Laplaian disibution help to find are non-diretional hanges beause they enhane linear features in any diretion in an image. They do not look at the gradient itself, but at the hanges in gradient. In their simplest form, they an be seen as the result of taking the seond derivative. In this step is mean of the blok or region S of image I having three olor hannel {red, green, blue}. So first is estimate whih at as the saling parameter of the laplae distribution. 1 I ( L) LS 1 P( ( L) S) LS 1 e 2 ( I M ( L) I ( L)) 3.4 Haze Thikness Estimation In this step olor adjustment parameter is alulate with the help of Laplae distribution values of eah blok. Here olor adjustment parameter is the ratio of the olor hromati parameter to the maximum value of the olor hromati parameter of eah hannel of the image. a max I Here is the olor hromati parameter obtain by the ratio of the maximum laplae distribution value of the olor region to the laplae distribution of the blok. Finally olor adjustment in the haze image is done by hange in pixel value ( I ( x) A ) J ( x) A max( t( x), t ) o P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 66

www.ijrar.om Here A is atmospheri light adjustment parameter in the blok so for eah blok it is evaluate by 3.4 Edge Combination A min(max( S) I t( x) min A Here edge obtain from initial image is ombine in the new edge position obtain after applying laplae and haze thikness steps. This involve following working: 1. Loop 1:m 2. Loop 1:n 3. If I(m,n) //edge pixel in old but not in proessed image. 4. I_new(m,n) I(m,n) 5. EndLoop 6. EndLoop 4. Experiment And Result In this setion, first introdue experimental settings, and then present the experimental results that validate the effetiveness of the approah. The experiments atually ontain two parts. This work is ompare with other previous work in [8] whih have utilize the laplae and haze thikness estimation only. 4.1 Data Sets In order to ondut the experiment an artifiial dataset whih is a olletion of images from different ategory are utilize. As images are of different format so first it is neessary to make it in readable format for experiment tool MATLAB. Now this olletion of images of different ategory is shown in table 1. Category Garden Jungle Animal Person Table 1. Dataset of Different ategory. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 67

www.ijrar.om 4.2 Results Original Image Image After Restortation Table 2. Results of various image from proposed restoration work. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 68

www.ijrar.om Images Visible Edge Restoration Proposed Previous Work Work 1 Garden 3.1503 2.6447 2 Jungle 3.9960 3.2902 3 Animal 4.4907 0.7239 4 Person 5.9658 0.6490 Table 3. Comparison of proposed work and previous work on visible edge restoration parameter. In table 3 It is obtained that proposed work is better as ompare to previous as edge restoration value of proposed work is higher as ompare to previous. So inlusion of edge feature in haze removal has inrease the performane of the work. Images Contrast Restoration Image Proposed Work Previous Work 1 Garden 7.7194 1.3662 2 Jungle 8.1391 3.3467 3 Animal 10.7419 7.7275 4 Person 9.4774-4.4575 Table 4. Comparison of proposed work and previous work on visible edge restoration parameter. In table 4 it is obtained that proposed work is better as ompare to previous as ontrast restoration image value of proposed work is higher as ompare to previous. So inlusion of edge feature in haze removal has inrease the performane of the work. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 69

www.ijrar.om Images Over or Under Exposed Metri Proposed Work Previous Work 1 Garden 2.7487 0.2838 2 Jungle 2.7253 0.2802 3 Animal 2.6916 0.0171 4 Person 2.6238 0.0030 Table 5. Comparison of proposed work and previous work on visible edge restoration parameter. In table 5 It is obtained that proposed work is better as ompare to previous as Over or Under Exposed Metri of Restoration image value of proposed work is higher as ompare to previous. So inlusion of edge feature in haze removal has inrease the performane of the work. 5. Conlusions A new ombination of laplae and edge feature is done in this work for dehazing image from different sene. The algorithm removes spatially varying haze based on the haze thikness estimation. As experiment is done on images of different environment and it is obtained that proposed work is better on all the evaluation parameters of de-hazing images. In Future improvements of the method will deal with possible orner, and histogram effets aused by the image proessing. REFERENCES [1] Z. Li, J. Zheng, and Z. Zhu, Contenet Adaptive Guided Image Filtering. In IEEE Int. Conf. Multimedia and Expo (Ime), 2014, 2014. [2] Liu, Feng, and Canmei Yang. "A Fast Method for Single Image Dehazing Using Dark Channel Prior." In Signal Proessing, Communiations And Computing (Isp), 2014 IEEE International Conferene On, Pp. 483-486. Ieee, 2014. [3] Alajarmeh, Ahmad, Rosalina Abdul Salam, Mohd Fadzli Marhusin, and Khairi Abdulrahim. "Real-time Video Enhanement for Various Weather Conditions Using Dark Channel and Fuzzy Logi." In Computer and Information Sienes (Ioins), 2014 International Conferene On, Pp. 1-6. IEEE, 2014. [4] Serikawa, Seiihi, and Huimin Lu. "Underwater Image Dehazing Using Joint Trilateral Filter." Computers & Eletrial Engineering 40.1 (2014): 41-50. [5] Hitam, M. S., W. N. J. H. W. Yussof, E. A. Awalludin, and Z. Bahok. "Mixture Contrast Limited Adaptive Histogram Equalization For Underwater Image Enhanement." In Computer Appliations Tehnology (Iat), 2013 International Conferene On, Pp. 1-5. IEEE, 2013. [6] A.Mittal, R. Soundararajan, And A.C. Bovik, Making A Completely Blind Image Quality Analyzer, Ieee Signal Proess. Lett., Vol. 20, No.3, Pp.209-212, Mar. 2013. [7] K. He, J. Sun, and X. Tang, Guided Image Filtering, Ieee Trans. Patt. Anal. Mah. Intell., Vol. 35, No. 6, Pp. 1397-1409, 2013. [8] Zhu, Qingsong, Shuai Yang, Pheng Ann Heng, And Xuelong Li. "An Adaptive And Effetive Single Image Dehazing Algorithm Based On Dark Channel Prior." In robotis and Biomimeti (Robio), 2013 IEEE International Conferene On, Pp. 1796-1800. IEEE, 2013. P r i y a S i n g h P a t e l & P r o f. M a n i s h a R a t h o r e Page 70