Narrow Results

In this paper, new operational definitions of binary morphological, both conditional and nonconditional, operations are proposed. The new operations are applied to detect boundary points from binary images. Comparisons of boundary detection algorithms using proposed, standard morphological, and gradient-based operations, showing the effectiveness of the proposed operations, are given. Comparative hardware implementations of standard and proposed morphological operations are also given. Main distinguishing aspects of the new operations are: speed and low hardware implementation (i.e., e.g., low number of buffers and D-Flip–Flops).

As an important identification certificate for citizens, ID card plays a significant role in daily life and its information has found its way into almost every aspect. However, traditional ways tend to adopt manual input, which is not only time-consuming and labor-intensive, but also expensive as well as inaccuracy. In this paper, we proposed a novel algorithm to locate and recognize ID card information, in which several fresh strategies are presented to rectify image, detect boundary, and locate information, respectively. To solve the problem of image rotating, the image is rectified by searching the best rotating angle that can lead to the maximum corner point projection peak. Meanwhile, the boundary of ID card is detected by finding the best lines in the predicted boundary area based on the deviation between the predicted boundary and the detected boundary, and the position of information is located by incorporating the prior information and the location relation between the key information. Experimental results show that the proposed algorithm can achieve a state-of-the-art effect for recognizing ID card’s information.

We have developed a generalized alphanumeric character extraction algorithm that can efficiently and accurately locate and extract characters from complex scene images. A scene image may be complex due to the following reasons: (1) the characters are embedded in an image with other objects, such as structural bars, company logos and smears; (2) the characters may be painted or printed in any color including white, and the background color may differ only slightly from that of the characters; (3) the font, size and format of the characters may be different; and (4) the lighting may be uneven.

The main contribution of this research is that it permits the quick and accurate extraction of characters in a complex scene. A coarse search technique is used to locate potential characters, and then a fine grouping technique is used to extract characters accurately. Several additional techniques in the postprocessing phase eliminate spurious as well as overlapping characters. Experimental results of segmenting characters written on cargo container surfaces show that the system is feasible under real-life constraints. The program has been installed as part of a vision system which verifies container codes on vehicles passing through the Port of Singapore.

Prostate boundary detection from ultrasound images plays an important role in prostate disease diagnoses and treatments. However, due to the low contrast, speckle noise and shadowing in ultrasound images, this still remains a difficult task. Currently, prostate boundary detection is performed manually, which is arduous and heavily user dependent. A possible solution is to improve the efficiency by automating the boundary detection process with minimal manual involvement. This paper presents a new approach based on the level set method to automatically detect the prostate surface from 3D transrectal ultrasound images. The user interaction in the initialization procedure is relieved by automatically putting the centroid of the initial zero level sets close to the image center. Region information, instead of the image gradient, is integrated into the level set method to remedy the "boundary leaking" problem caused by gaps or weak boundaries. Moreover, to increase the accuracy and robustness, knowledge-based features, such as expected shape (kidney-like) and ultrasound appearance of the prostate (looking from within the gland, the intensities are transitions from dark to light), are also incorporated into the model. The proposed method is applied to eight 3D TRUS images and the results have shown its effectiveness.

Background: Breast cancer is a common and dreadful disease in women. One in five cancers in Singaporean women is due to breast cancer. Breast health is every woman's right and responsibility. In average, every $100 spent on breast mammogram screening, an additional $33 was spent on evaluating possible false-positive results. Thermography, with its non-radiation, non-contact and low-cost basis has been demonstrated to be a valuable and safe early risk marker of breast pathology, and an excellent case management tool available today in the ongoing monitoring and treatment of breast disease. The surface temperature and the vascularization pattern of the breast could indicate breast diseases and early detection saves lives. To establish the surface isotherm pattern of the breast and the normal range of cyclic variations of temperature distribution can assist in identifying the abnormal infrared images of diseased breasts. Before these thermograms can be analyzed objectively via computer algorithm, they must be digitized and segmented. The authors present a method to segment thermograms and extract useful region from the background. Thermography could detect the presence of tumors much earlier and of much smaller size than mammography. This paper thus aims to develop an intelligent diagnostic system based on thermography for the detection of tumors in breast. Methods: We have examined about 50 normal, healthy female volunteers in Nanyang Technological University and 130 patients in Singapore General Hospital. We did the examinations for some of them continuously for two months. From these examinations, we obtained about 1000 thermograms for contact and 800 thermograms for non-contact approaches. Standard ambient conditions were observed for all examinations. The thermograms obtained were analyzed. The first step in processing these thermograms is image segmentation. Its aim is to discern the useful region from the background. In general, autonomous segmentation is one of the most difficult tasks in image processing. This step in the process determines the eventual success or failure of the analysis. In this work, two different techniques have been presented to extract the objects from the background. Results: After analyzing these thermograms and with reference to some relevant well-documented papers, we were able to classify the thermograms. The step is very useful in identifying the normal or suspected (abnormal) thermograms. A series of thermograms was studied with the help of the in-house developed computer software. On the basis of the anatomic and vascular symmetry, the surface temperature distributions of both left and right breasts were compared. The surface isotherm pattern of breasts can indicate the local metabolism and vascularity of the underlying tissues, and the change in local blood or glandular activities can be reflected in the surface temperature of breast. We evaluated the temperature distribution pattern and the menstrual cyclic variation of temperature with time. All these results can be used to detect breast cancer. Conclusion: Automatic identification of object and surface boundary of breast thermal images is a difficult and challenging task. Both the traditional snake and gradient vector flow snake failed to detect the boundary of these images successfully. In this work, a new method is proposed in conjunction with image pre-processing, image transition, image derivative, filtering and gradient vector flow snake. This novel method can easily detect the boundary of the breast thermal image with good agreement.

To develop a high-quality TTS system, an appropriate segmentation of continuous speech into the syllabic units plays a vital role. The significant objective of this research work involves the implementation of an automatic syllable-based speech segmentation technique for continuous speech of the Hindi language. Here, the parameters involved in the segmentation process are optimized to segment the speech syllables. In addition to this, the proposed iterative splitting process containing the optimum parameters minimizes the deletion errors. Thus, the optimized iterative incorporation can discard more insertions without merging the frequent non-iterative incorporation. The mixture of optimized iterative and iterative incorporation provides the best accuracy with the least insertion and deletion errors. The segmentation output based on different text signals for the proposed approach and other techniques namely GA, PSO and SOM is accurately segmented. The average accuracy obtained for the proposed approach is high with 97.5% than GA, PSO and SOM. The performance of the proposed algorithm is also analyzed and gives better-segmented accuracy when compared with other state-of-the-art methods. Here, the syllable-based segmented database is suitable for the speech technology system for Hindi in the travel domain.

The prevalence of ovarian tumor malignancy can be monitored by the degree of irregularity in the ovarian contour and by the septal structure inside the tumor observed in ultrasonic images. However the 2D ultrasonic images can not integrate 3D information form the ovarian tumor. In this paper, we present an algorithm that can render the 3D image of an ovarian tumor by reconstructing the 2D ultrasonic images into a 3D data set. This is based on sequentially boundary detection in a series of 2D images to form a 3D tumor contour. This contour is then used as a barrier to remove the data containing the other tissue adhering to the tumor surface. The final 3D image rendered by the isolated data provides a clear view of both the surface and inner structure of the ovarian tumor.

As medical imaging techniques have been developed, efficient manipulation and visualization of the obtained images are important topics to improve diagnostic accuracy and to expand their applications. In order to use the images effectively, images have to be processed to meet the need of the application. These processes include image segmentation and 3D visualization. Here, we describe various segmentation techniques such as boundary-based and region-based segmentation, and visualization techniques such as surface rendering and volume rendering, together with visualization of functional MRI and diffusion tensor MRI of brain.

This chapter describes image processing methods for document image analysis. The methods are grouped into four categories, namely, image acquisition, image transformation, image segmentation, and feature extraction. In image acquisition, we describe the process of converting a document into its numerical representation, including image coding as a means to reduce the storage requirement. Image transformation addresses image-to-image operations, which comprise a large spectrum of techniques ranging from geometrical correction, filtering and figure-background separation to boundary detection and thinning. In image segmentation, we describe four popular techniques, namely, connected component labeling, X-Y-tree decomposition, run-length smearing, and Hough transform. Finally, a number of feature extraction methods, which constitute the basis of image classification, are presented.

The recent surge of Tuberculosis (TB) cases in India and in its seven North-Eastern States in particular, has introduced a renewed interest in the estimation of TB risk surfaces in the neighborhood and identification of units having elevated risk. The present paper maps the TB risk surfaces on the basis of district level incidence rates for the seven neighboring North-eastern states of India. The risk surface is represented with a set of random effects through Bayesian Hierarchical approach. In the present case, the random effects are modeled through conditional autoregressive (CAR) prior exhibiting a single level of spatial smoothness. Also, attempt has been made to identify the risk boundaries having elevated risk through localized spatial structure by modeling the weighted contiguity matrix for geographically adjacent areas as binary random quantities.