M. K. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Trans. on Consum. Electron.,vol. 42, pp. 557--565, Aug 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... have been applied to image retrieval in recent years: boosting [136] clustering [99] edge extraction [54] grouping [36] hidden Markov models [82] various histograms [33, 133, 107] image segmentation [3, 20, 72, 146] invariant features [111, 64, 41, 27] keypoint extraction [111] moments [29, 74], motion estimation [2] probabilistic matching [1, 140, 139] self organizing maps [67] shape matching [64, 68] texture features [33, 70, 43] transportation problem solving [101] and wavelets [59, 74] just to mention some. For a more extensive overview refer to the survey papers [102, 128] ....

.... [3, 20, 72, 146] invariant features [111, 64, 41, 27] keypoint extraction [111] moments [29, 74] motion estimation [2] probabilistic matching [1, 140, 139] self organizing maps [67] shape matching [64, 68] texture features [33, 70, 43] transportation problem solving [101] and wavelets [59, 74], just to mention some. For a more extensive overview refer to the survey papers [102, 128] 142] gives an overview too but focuses on comparing existing systems. 2.1 Online versus offline systems The approaches considered so far for content based image retrieval can be classified as one of ....

[Article contains additional citation context not shown here]

M. K. Mandal, T. Aboulnasr, and S. Panchanathan. Image indexing using moments and wavelets. IEEE Transactions on Consumer Electronics, 42(3):557-- 565, 1996.

....a very compact representation that required minimal processing to support small mobile robots. Color is a very important cue in extracting information from an image, and color histogram comparison has recently become a popular technique for image and video indexing[SB91] SO95] LD95] NM95] Pan96] The popularity (c) d) e) f) a) b) c) d) e) f) Fig. 7. Left: Images from six different rooms. Right: Images taken in one room at different times. of color as an index resides in its ease of computation and effectiveness[LM97] Some papers suggest that color histograms are ....

M. K. Mandal T. Abdulnasir S. Panchanathan. Image indexing using moments and wavelets. IEEE Transactions on Consumer Electronics, 42(3):45--48, August 1996.

....distribution obtained by projecting it onto a PCA base. Using Funt et al. s dataset of 11 objects photographed under 4 different illuminants they have demonstrated how removing the mean from the log chromaticity coordinates removes the effect of illuminant variance on the color distribution. In [46] Mandal et al. have presented a method to define illuminant invariant moments. Assuming that the change in illumination is uniform, and the illumination does not produce a nonlinear effect on the image, the histograms of an image with varying lighting conditions can be approximated as a translated ....

Mandal M. K., T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets", IEEE Trans. Consumer Electron. Vol. 42(3), pp. 557-565, 1996.

....carry some redundant comparison. Ideally the information that has been compared at the lower resolution search should not be re compared again at the higher resolution. Orthogonal moments Some retrieval performance gain is expected if we replace regular moments with orthogonal Legendre moments [27]. Since Legendre moments are orthogonal, the 37 38 CHAPTER 6. OPEN QUESTIONS AND FUTURE WORK similarity metric based on Euclidian distance would provide superior indexing performance. However whether we can retain the invariant properties for the orthogonal moments is an open question. Better ....

M.K. Mandal, T. Aboulnasr, and S. Panchanathan. Image indexing using moments and wavelets. IEEE Trans. on Consumer Electronics, 42:557--565, August 1996.

.... into two approaches, structural and statistical, both of them lack the human visual constraints [1] Recently, since a multi resolution paradigm with a pyramid structure matches well to human texture perception, a number of researchers introduced this structure to the retrieval applications [2][3]. Unfortunately, the wavelet transform lacks the translation and rotation invariant properties. This results in the mismatch of the retrieval process when the image orientation is altered. To overcome this problem, the combination of wavelets and moments is proposed [3] Alternatively, steerable ....

....retrieval applications [2] 3] Unfortunately, the wavelet transform lacks the translation and rotation invariant properties. This results in the mismatch of the retrieval process when the image orientation is altered. To overcome this problem, the combination of wavelets and moments is proposed [3]. Alternatively, steerable pyramid structure, which has the translation and rotation invariant properties is also introduced [1] However, this algorithm works well only when the images have the same luminance intensity. To incorporate all the invariant properties, we investigated the statistical ....

[Article contains additional citation context not shown here]

M. K. Mandal, T. Aboulnasr and S. Phanchanathan, " Image indexing using moments and wavelets," IEEE Trans. on Consumer Electronics, vol. 42, pp. 557-565, Aug. 1996.

....been classi ed into classes a priori through some viable means, and (ii) that multiple prototypes are available per class. A novel distance metric is de ned to take advantage of these assumptions to improve the performance, with an image representation based on Gabor and wavelet features [3] 4] [5]. In contrast to the NN type metrics, the NFL metric makes use of the available information about classes contained in the multiple prototypes of each class. A subspace is constructed out from the whole feature space for each image class, based on the prior knowledge of multiple prototypes to ....

M. K. Mandal, T. Aboulnasr, and S. Panchanathan, Image indexing using moments and wavelets, IEEE Transactions on Consumer Electronics, vol. 42, no. 3, pp. 557-565, 1996.

....and implementation of image database systems supporting queries by image content are image feature extraction[7, 10, 13, 15, 17] image content representation and organisation of stored information[42, 43] CHAPTER 2. PROBLEM FORMULATION AND CONTRIBUTIONS 10 searching and retrieving strategies[6, 25, 40, 41], and user interface design. Addressing such issues has become the object of intensive research activities in many areas of computer science over the past few years. Advances mainly in the areas of databases and computer vision research have resulted in methods which can be used for image ....

....using moments, but I do not discuss it here[17] 30] It sounds that using moments can solve the image alignment problem easily. However, it is noise sensitive. It only applies to images whose variation is very small. 3. 3 Wavelet Review There are many publications on wavelet analysis[39] [40]. A short and clear theory can be found in [39] Wavelets are functions generated from one single function Psi by dilations and translations Psi a;b (t) jaj Gamma1=2 Psi( t Gamma b a ) 3.35) The mother wavelet Psi has to satisfy the condition Z Psi(x)dx = 0 (3.36) CHAPTER 3. ....

[Article contains additional citation context not shown here]

M. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets", IEEE Trans. on Consumer Electronics, V.42, #3, pp. 557-565, 1996.

....the difficulties of the pure annotation based approach, since the feature extraction process can be made automatic and the image s own content is always consistent. Since its advent, CBIR has attracted great research attention, ranging from government [4, 5] industry [2, 6, 7] to universities [8, 9, 10, 11, 12]. Even ISO IEC has recently launched a new work item, MPEG7 [13, 14, 15, 16] to define a standard Multimedia Content Description Interface. Many special issues from leading journals have been dedicated to CBIR [17, 18, 19, 20] and many CBIR systems, both commercial [1, 2, 3, 6, 7] and academic ....

....12] Even ISO IEC has recently launched a new work item, MPEG7 [13, 14, 15, 16] to define a standard Multimedia Content Description Interface. Many special issues from leading journals have been dedicated to CBIR [17, 18, 19, 20] and many CBIR systems, both commercial [1, 2, 3, 6, 7] and academic [8, 9, 10, 11, 12], have been developed recently. Despite the extensive research effort, the retrieval techniques used in CBIR systems lag behind the corresponding techniques in today s best text search engines, such as Yahoo, Alta Vista, Lycos, etc. At the early stage of CBIR, research primarily focused on ....

[Article contains additional citation context not shown here]

M. K. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Transactions on Consumer Electronics, vol. 42, pp. 557--565, Aug 1996.

....comparing the sign of the most significant wavelet coefficients. In Liang et al. 14] a wavelet based image retrieval system is developed, where the translation variance of wavelet transform is combatted by adopting the number of significant coefficients in each subband as the primary feature. In [15], histogram comparison of highpass subbands is proposed by modeling the histogram of each subband by generalized Gaussian distribution and the variance and shape parameters are used as features. As a summary, we deduce that only some of the proposed algorithms use scalable features [14] ....

M.K. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Transactions on Consumer Electronics, vol. 42, no. 3, pp. 557--565, 1996.

....of the pure annotation based approach, since the feature extraction process can be automated, while ensuring that the image content is always consistent . Since its advent, CBIR has attracted significant research attention, including government [10, 11] industry [2, 4, 8] and academia [9, 12, 13, 20, 31]. In fact, ISO IEC has launched a new work item, MPEG 7 [16, 17, 18] to define a standard Multimedia Content Description Interface. Many CBIR systems, both commercial [2, 4, 6, 7, 8] and academic [9, 12, 13, 20, 31] have been developed recently. Despite this extensive research effort, there ....

....attention, including government [10, 11] industry [2, 4, 8] and academia [9, 12, 13, 20, 31] In fact, ISO IEC has launched a new work item, MPEG 7 [16, 17, 18] to define a standard Multimedia Content Description Interface. Many CBIR systems, both commercial [2, 4, 6, 7, 8] and academic [9, 12, 13, 20, 31], have been developed recently. Despite this extensive research effort, there remain many challenges to be addressed before a successful image retrieval system becomes more viable and practical: Region based techniques vs. entire image based techniques. Most existing systems only support queries ....

M. K. Mandal, T. Aboulnasr, and S. Panchanathan. Image Indexing Using Moments and Wavelets. IEEE Transactions on Consumer Electronics, 42(3):557--565, August 1996.

....of the pure annotation based approach, since the feature extraction process can be automated, while ensuring that the image content is always consistent . Since its advent, CBIR has attracted significant research attention, including government [14, 15] industry [5, 7, 10] and academia [13, 17, 18, 23, 35]. In fact, ISO IEC has launched a new work item, MPEG 7 [1, 2, 3] to define a standard Multimedia Content Description Interface. Many CBIR systems, both commercial [5, 7, 21, 9, 10] and academic [13, 17, 18, 23, 35] have been developed recently. Despite this extensive research effort, there ....

....attention, including government [14, 15] industry [5, 7, 10] and academia [13, 17, 18, 23, 35] In fact, ISO IEC has launched a new work item, MPEG 7 [1, 2, 3] to define a standard Multimedia Content Description Interface. Many CBIR systems, both commercial [5, 7, 21, 9, 10] and academic [13, 17, 18, 23, 35], have been developed recently. Despite this extensive research effort, there remain many challenges to be addressed before a successful image retrieval system becomes more viable and practical: ffl Most existing systems only support queries based on entire images. However, more often than not, ....

M. K. Mandal, T Aboulnasr, and S. Panchanathan. Image indexing using moments and wavelets. IEEE Transactions on Consumer Electronics, 42(3):557--

....First, indexing can be done hierarchically exploiting the multiresolution property of DWT. Secondly, edge and shape of objects can be estimated easily in DWT domain. Finally, directional information from various directional subbands can be employed to enhance indexing performance. Mandal et al. [10] have proposed an indexing technique based on histogram of wavelet coefficients of different bands. Liang et al. 11] have proposed a joint image coding and indexing in wavelet domain where wavelet packet tree structure and subband significance are employed as features for indexing. Wang et al. ....

M. K. Mandal, T. Aboulnasr and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Trans. on Consumer Electronics, Vol. 42, No. 3, Aug 1996.

....corresponding to the least histogram difference is retrieved for visual interpretation. Stricker et al. 8] have proposed a reduced complexity technique that compares a few of the histograms moments instead of the actual histograms. Mandal et al. Journal of Electrionic Imaging, April 1998 2 [9] have shown that orthogonal Legendre moments provide superior indexing performance compared to regular or central moments. With the advent of various image compression standards [10,11] the current and future databases are likely to employ compression techniques for efficient storage. This has ....

....directly on the compressed data (see Fig. 2) Various indexing techniques have been proposed employing KarhunenLoeve transform (KLT) 13] discrete cosine transform (DCT) 14] and discrete wavelet transform (DWT) 15] A technique exploiting the directional property of DWT has been proposed in [9]. Here, the histograms of different wavelet bands of the query image are employed as an index. The indexing techniques proposed in the literature generally assume that similar images in the database have the same brightness. This assumption is violated in practice due to the following reasons: i) ....

[Article contains additional citation context not shown here]

M. K. Mandal, T. Aboulnasr and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Transactions on Consumer Electronics 42(3), 557-565, (Aug. 1996).

No context found.

M. K. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets," IEEE Trans. on Consum. Electron.,vol. 42, pp. 557--565, Aug 1996.

No context found.

M. K. Mandal, T. Aboulnasr, and S. Panchanathan, "Image indexing using moments and wavelets", IEEE Trans. Consumer Electronics, 42(3):557-565, 1996.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC