49_ by ides editor

Proc. of Int. Conf. on Advances in Computer Science 2010

Morphological Background Detection and Enhancement of Color Images with Poor lighting A.Vamsi Krishna, Ch. Pavan Teja, Dr. A.Sri Krishna, Dr. B. Raveendra Babu, vamsikrishna.it@gmail.com, pavanteja.ch@gmail.com, atlurisrikrishna@yahoo.com , rbhogapathi@yahoo.com Department of Computer Science and Information Technology, R.V.R & J.C College of Engineering, Chandramoulipuram, Guntur-19. Abstract- The contrast enhancement is an important visualization technique in digital image processing to improve the quality of an image. The most common technique in image processing is to enhance the contrast with poor lighting. A new morphological operator, opening by reconstruction, is defined to enhance and normalize the contrast in color images with poor lighting. Contrast enhancement is carried out by the application of two operators based on Weber’s law notion. The performance of operators is illustrated through the processing of images with different backgrounds. Index terms- Morphological Filters, Weber’s Ratio, Contrast Enhancement.

I. INTRODUCTION The contrast enhancement problem in digital images can be approached from various methodologies, among which is mathematical morphology (MM). Initial studies on contrast enhancement in this area were carried out by Meyer and Serra [1], who introduced the contrast mappings notion. Such operators consist in accordance to some proximity criterion, in selecting for each point of the analyzed image, a new grey level between two patterns (primitives) [1]. Other works based on the contrast mapping concept have been developed in [2]–[4]. With regard to MM, several studies based on contrast multiscale criterion have been carried out [5]–[7]. In the work proposed by Mukhopadhyay and Chanda [6], a scheme is defined to enhance local contrast based on a morphological top hat transformation. Kasperek [7] implemented a processing system in real time for its application in the enhancement of angiocardiographic images, based on the work carried out by Mukhopadhyay. There are techniques based on data statistical analysis, such as global and local histogram equalization. During the histogram equalization process, grey level intensities are reordered within the image to obtain a uniform distributed histogram [8],[12],[13]. However, the main disadvantage of histogram equalization is that the global properties of the image cannot be properly applied in a local context [9], frequently producing a poor performance in detail preservation. In [10], the authors apply the proposed operators to some images with poor lighting with good results. The paper is organized as follows. Section II presents a brief background on some morphological transformations and Weber’s law. Section III introduces opening by reconstruction transformations that enhance images with © 2010 ACEEE DOI: 02.ACS.2010.01.49

200

poor lighting. Section IV includes different results and discussions. Finally, conclusions are presented in Section V. II. BACKGROUND A. Morphological transformations: In binary morphological image analysis, a 2D image is defined as a subset of the 2-D Euclidean space R×R or its digitized equivalent Z×Z. In this paper, we deal only with digital images that are defined as subsets of Z×Z. For an image A⊆ Z×Z and a point u ε Z×Z, the transition of A by u is defined by equation (1) (1) The two most fundamental morphological operations dilation and erosion are defined by equations (2) and (3) respectively (2) (3) where B is a structuring element. Another important pair of morphological operations are opening and closing. They are defined in terms of dilation and erosion, by equations (4) and (5) respectively (4) (5) Weber’s Law In psycho-visual studies, the contrast C of an object with luminance ‘Lmax’ against its surrounding luminance ‘Lmin’ is defined as follows [14]: C= (Lmax – Lmin) / Lmin

(6)

If L = ’Lmin’ & ∆L =Lmax – Lmin Then C = ∆L / L

(7)

Indicates that ∆ (logL) is proportional to C; therefore, Weber’s law can be expressed as: C = klogL + b (8) Where k and b are constants, being the background. In our case, an approximation to Weber’s law is considered by taking the luminance L as the grey level intensity of a function f (image); in this way, expression (5) is written as

Proc. of Int. Conf. on Advances in Computer Science 2010

C=klog f + b Where C is the contrast of the image.

(9)

III. METHODOLOGY The algorithm works in stages the first part identifies the background of the images and then image enhancement takes place. The back ground criterion is obtained by considering the maximum and minimum intensity values in each sub image in the case of analysis by blocks. In the case of the opening by reconstruction the background intensity is identified by means of morphological operations which can be observed by taking global properties in to the consideration. Algorithm 1: Image Background Approximation by Blocks (IBAB): Step 1: Input a color image Step 2: The original image is divided into sub images based on the intensity values. Step 3: The minimum and maximum intensities in each sub image are calculated as given in equations (10) and (11) (10)

Algorithm 2: Image Background Determination Using the Opening By Reconstruction (IBDOR): Step 1: Input a color image Step 2: The background criteria for total image is calculated in eq.(15) Background criteria (15) Step 3: Image background which gives the local information of the image is computed using eq (16) (16) where the structuring element of size Step 4: The following expression derived from (8) is proposed to enhance the contrast in images with poor lighting is given in eq (17) (17) and

Where maxint=255. Step 5: Stop.

(11) Step 4: The background criteria using maximum equation (12)

and minimum

is calculated by values as given in Fig 1(a)

Fig 1(b)

(12) Step 5: For the blocks division the expression is obtained as

Fig 1(c) Fig 1 :(a) Original image (b) Enhanced image after applying IBAB c) Enhanced images after applying IBDOR

(13)

where v(x) indicates the background of the image. Step 6: The background criteria, obtained is used to find enhancement in the proposed expression as given in equation (14)

(14) The constant

Fig2(a)

Fig2(b)

Fig 2 :(a) Original images and (b) Enhanced images after applying IBDOR

in (14) is obtained as follows:

With Where mi is the minimum and Mi is the maximum intensities. Step 7: Stop

ÂŠ 2010 ACEEE DOI: 02.ACS.2010.01.49

201

Fig3(a)

Fig3(b)

Fig 3 :(a) Original images and (b) Recognizing objects using IBDOR

Proc. of Int. Conf. on Advances in Computer Science 2010

REFERENCES

Fig4(a)

Fig4(b)

Fig 4 :(a) Original image and b) over illuminated enhanced image after applying IBDOR

IV. RESULTS AND DISCUSSIONS The two methodologies to enhance the contrast have been applied on different color images taken with different illumination. It has been observed that enhanced image using IBAB gives good contrast but generates additional patches at the brightest region of an image as shown in fig 1(b). It has been observed that IBDOR algorithm results in enhanced image with good contrast as shown in fig 1(c). The patches generated using IBAB algorithm has been taken care in the case of IBDOR algorithm. This algorithm works efficiently in the case of poor lighting. This is clearly evident from fig 3(a) & (b). The advantage part of the algorithm is, we can easily identify hidden alphabets or numbers on vehicle number plates. The disadvantage part of IBAB is that it takes more time comparative to IBDOR. The algorithms have limitation of over illumination. This can be observed in fig (2) where the text part is lost in an image. The algorithm is basically used on the color images with poor lighting which helps us identifying those objects which are completely invisible in the images. It has been observed that over illumination is also observed on some part of the image while other part is enhanced as shown in fig (4). V. CONCLUSION This paper considers the background of an image and then enhances the image. Two methods are proposed the first method IBAB has been applied on different images and has been observed that new patches are introduced and time complexity is more. These limitations are eliminated in the case of IBDOR. In both IBAB and IBDOR methods over illumination is observed in the enhanced images. Future work involves in the elimination of this over illumination.

in Annotationes anatomicae et physiologicae.Leipzig, Germany: Koehler, 1834

ACKNOWLEDGEMENT

[17] S.-D. Chen and A. Ramli, “Minimum mean brightness

The authors would like to express their gratitude to the Management and Principal R.V.R & JC College of Engineering, Chowdavaram, Guntur for providing necessary research infrastructure for carrying out this work.

[1]. F. Meyer and J. Serra, “Contrast and Activity Lattice,” Signal Process., vol. 16, pp. 303–317, 1989. [2]. I. R. Terol-Villalobos, “Morphological image enhancement and segmentation,” in Advances in Imaging and Electron Physics, P. W. Hawkes, Ed. New York: Academic, 2001, pp. 207–273. [3]. I. R. Terol-Villalobos, “Morphological connected contrast mappings based on top-hat criteria: A multiscale contrast approach,” Opt. Eng., vol. 43, no. 7, pp. 1577–1595, 2004. [4]. J. D. Mendiola-Santibañez and I. R. Terol-Villalobos, “Morphological contrast mappings based on the flat zone notion,” Computación y Sistemas, vol. 6, pp. 25– 37, 2002. [5]. A. Toet, “Multiscale contrast enhancement with applications to image fusion,” Opt. Eng., vol. 31, no. 5, 1992. [6]. S. Mukhopadhyay and B. Chanda, “A multiscale morphological approach to local contrast enhancement,” Signal Process., vol. 80, no. 4, pp. 685–696, 2000. [7]. J. Kasperek, “Real time morphological image contrast enhancement in virtex FPGA,” in Lecture Notes in Computer Science. New York: Springer, 2004. [8]. C. R. González and E.Woods, Digital Image Processing. Englewood Cliffs, NJ: Prentice Hall, 1992. [9]. R. H. Sherrier and G. A. Johnson, “Regionally adaptive histogram equalization of the chest,” IEEE Trans. Med. Imag., vol. MI-6, pp. 1–7, 1987. [10]. A. Majumder and S. Irani, “Perception-based contrast enhancement of images,” ACM Trans. Appl. Percpt., vol. 4, no. 3, 2007, Article 17. [11]. Z. Liu, C. Zhang, and Z. Zhang, “Learning-based perceptual image quality improvement for video conferencing,” presented at the IEEE Int. Conf. Multimedia and Expo (ICME), Beijing, China, Jul. 2007. [12]. J. Serra and P. Salembier, “Connected operators and pyramids,” presented at the SPIE. Image Algebra and Mathematical Morphology, San Diego, CA, Jul. 1993. [13]. P. Salembier and J. Serra, “Flat zones filtering, connected operators and filters by reconstruction,” IEEE Trans. Image Process., vol. 3, no. 8, pp. 1153– 1160, Aug. 1995. [14]. E. Peli, “Contrast in complex images,” J. Opt. Soc. Amer., vol. 7, no. 10, pp. 2032–2040, 1990. G. de Haan, in Video Processing for Multimedia Systems, Eindhoven, The Netherlands, 2000. [15]. G. de Haan, in Video Processing for Multimedia Systems, Eindhoven, The Netherlands, 2000. [16] E. H. Weber, “De pulsui, resorptione, audita et tactu,”

202

error bi-histogram equalization in contrast enhancement,” IEEE Trans. Consum. Electron., vol. 49, no. 4, pp. 1310– 1319, Apr. 2003.

Proc. of Int. Conf. on Advances in Computer Science 2010

BIODATA Anne Vamsi Krishna is pursuing his Graduation in Information Technology, in R.V.R & J.C college of Engineering , chowdavarm,Guntur, Andhra Pradesh, India. His enthusiastic attitude bought for him prizes in Project presentations and coding competitions in various national level inter-collegiate competitions. Besides, this he has been actively participating and presenting papers in student technical symposium seminars at national level.His area of interest includes Image processing, pattern recognisition and cloud computing. Chirumamilla Pavan Teja is pursuing his Graduation in Information Technology, in R.V.R & J.C college of Engineering , chowdavarm,Guntur, Andhra Pradesh, India. He secured many prizes in various software designing activities in national level inter-collegiate competitions. In addition to this he has been actively participating and presenting papers in student technical symposium seminars at national level. His area of interest includes Image processing, pattern recognisition.

ÂŠ 2010 ACEEE DOI: 02.ACS.2010.01.49

203

A.Srikrishna received the AMIE degree in Electronics & Communication Engineering from Institution of Engineers; Kolkatta in 1990, M.S degree in Software Systems from Birla Institute of Technology and Science, Pilani in 1994, M.Tech degree in Computer Science from Jawaharlal Nehru Technological University (JNTU) in 2003. She has completed her Ph.D from JNTUK in Computer Science and Engg. She has worked as lecturer from (1991-95).She is working in RVR &JC College of Engineering, Guntur from 15 years as Assistant Professor , Associate Professor and Professor. Her research interest includes Image Processing and Pattern Recognition. She is Associate member of IE (I) and member of CSI. B. Raveendra Babu received the Ph.D degree from S.V.University, Tirupati, M.S degree in Software Systems from Birla Institute of Technology and Science, Pilani, M.Tech degree in Computer S cience from Anna University. He has 24 years of teaching experience as Assistant Professor, Associate Professor, Professor and presently he is working as Head, Dept of Computer Science and Engineering at RVR &JC College of Engineering, Guntur. He is a life member for CSI and ISTE. He has published more than 20 research publications in various National, Inter National conferences, proceedings and Journals.