Fig 1 - uploaded by Anwar Ali Yahya
Content may be subject to copyright.
Source publication
The difficulties of locating a desired facial image in a large and varied collection are now the current main problem in this field. In order to search in such a large and varied images’ collection, there is a growing need for efficient storage and retrieval techniques. In this research, an effective approach towards content-based human facial imag...
Contexts in source publication
Context 1
... i(x,y) represents the original image and ii(x,y) is the integral image as shown in Fig.1. Four kinds of rectangles features are used with varying numbers of sub- rectangle: a tow horizontal rectangle features, a two vertical rectangle feature, a three-rectangle feature and a four- rectangle feature, as shown in Fig. 1.Within any image sub- window, the total number of Haar-like features, that will be generated based on the integral image method and the four rectangles features, is very huge. ...
Context 2
... i(x,y) represents the original image and ii(x,y) is the integral image as shown in Fig.1. Four kinds of rectangles features are used with varying numbers of sub- rectangle: a tow horizontal rectangle features, a two vertical rectangle feature, a three-rectangle feature and a four- rectangle feature, as shown in Fig. 1.Within any image sub- window, the total number of Haar-like features, that will be generated based on the integral image method and the four rectangles features, is very huge. The feature set should be reduced to a small number of important features. AdaBoost learning method [12] was successfully employed to select a restricted number ...
Context 3
... and Jones arranged the classifiers in cascade architecture as shown in Fig. 1. In the cascade architecture, a series of classifiers are applied to every sub-window. Negative sub-windows will be rejected and the positive ones sub-windows will be detected and selected in the beginning stages by the initial classifier with fewer features and less computational time. The cascade classifiers in the final stages with ...
Context 4
... queried image. Unlike the eigenfaces features, performance of the color histogram algorithm somewhat depends on the dimension of the features vectors. With color histogram, increasing the size of the bines, would result in slight increase in the dimension of the features vectors, leading to improved retrieval performance. Results in Table 6 and Fig. 10 on the local database show the accuracies of the facial image retrieval system based on color histogram features with different sizes of ...
Context 5
... this influence is clear on color images database, as it is shown in Table 7 and Fig. 11 the distribution of the color space coordinate has no influence on gray level image database. This is because the three channels of the gray image carry the same information. The colour histogram algorithm conducts image colour analysis without consideration for locations of colour components in the image. Consequently, object location ...
Context 6
... examples of facial retrieval based on the color histogram algorithm using the proposed method of facial image segmentation have been provided. Fig. 12 and Fig. 13 respectively show the results of visual query and image retrieval of the color histogram using the local database. The recall method of performance measure shows that 100% accuracy was achieved for the top 10, 16, and 25 cut-off levels, where all the 10 images related to the query image were retrieved in the first and ...
Context 7
... examples of facial retrieval based on the color histogram algorithm using the proposed method of facial image segmentation have been provided. Fig. 12 and Fig. 13 respectively show the results of visual query and image retrieval of the color histogram using the local database. The recall method of performance measure shows that 100% accuracy was achieved for the top 10, 16, and 25 cut-off levels, where all the 10 images related to the query image were retrieved in the first and second rows of ...
Similar publications
Image features play a vital role in image retrieval. This chapter presents the use of Zernike moment features for retrieving the binary and gray level images from established image databases. To retrieve a set of similar category of images from an image database, up to 25 Zernike moment features from order zero to order 8 were utilized and experime...
Image low-level information, such as color, texture, and shape, were generally dealt with separately and combined together gruffly. Their image-describing effect in image retrieval was weakened. This paper determines and extracts one group of texture elements from images to mainly express the image texture information and, during this procedure, th...
Query images presented to content-based image retrieval systems often have various different interpretations, making it difficult to identify the search objective pursued by the user. We propose a technique for overcoming this ambiguity, while keeping the amount of required user interaction at a minimum. To achieve this, the neighborhood of the que...
We present a probabilistic model for Sketch-Based Image Retrieval (SBIR) where, at retrieval time, we are given sketches from novel classes, that were not present at training time. Existing SBIR methods, most of which rely on learning class-wise correspondences between sketches and images, typically work well only for previously seen sketch classes...
This article presents a novel approach to training classifiers for concept detection using tags and a variant of Support Vector Machine that enables the usage of training weights per sample. Combined with an appropriate tag weighting mechanism, more relevant samples play a more important role in the calibration of the final concept-detector model....
