Measurable properties of B-mode ultrasound images of thyroid gland in subjects with oncological diseases were compared in this survey. Data from 12 patients with breast cancer and 8 with colorectal cancer were analyze...
详细信息
ISBN:
(纸本)0889864349
Measurable properties of B-mode ultrasound images of thyroid gland in subjects with oncological diseases were compared in this survey. Data from 12 patients with breast cancer and 8 with colorectal cancer were analyzed in the study. Raw gray values of individual image pixels and spatial texture features were used as quantitative image features and were compared between both groups. The possible relation between quantitative ultrasound image features and serological markers of thyroid autoimmunity was analyzed. Significant correlation between serum levels of antibodies to thyroid peroxidase and image texture features was found in subject with breast cancer. On contrary, similar relation was not found in subjects with colorectal cancer. This finding shows that the high incidence of positivity of antibodies to thyroid peroxidase in patients with breast cancer is accompanied with changes in ultrasound image in thyroid gland. This work is one of the first practical applications of quantitative indicators of ultrasound image (texture features) in clinical medicine.
When taking pictures, professional photographers apply photographic composition rules, e.g. rule of thirds. The rule of thirds says to place the main subject's center at one of four places: at 1/3 or 2/3 of the pi...
详细信息
ISBN:
(纸本)0819452041
When taking pictures, professional photographers apply photographic composition rules, e.g. rule of thirds. The rule of thirds says to place the main subject's center at one of four places: at 1/3 or 2/3 of the picture width from left edge, and 1/3 or 2/3 of the picture height from the top edge. This paper develops low-complexity unsupervised methods for digital still cameras to (1) segment the main subject and (2) realize the rule-of-thirds. The main subject segmentation method uses the auto-focus filter, opens the shutter aperture fully, and segments the resulting image. These camera settings place the main subject in focus and blur the rest of the image by diffused light. The segmentation utilizes the difference in frequency content between the main subject and blurred background. The segmentation does not depend on prior knowledge of the indoor/outdoor setting or scene content. The rule-of-thirds method moves the centroid of the main subject to the closest of the four rule-of-thirds locations. We first define an objective function that measures how close the main subject placement obeys the rule-of-thirds, and then reposition the main subject in order to optimize the objective function. For multiple main subjects, the proposed algorithm could be extended to use rule-of-triangles by adding an appropriate constraint.
A programmable parallel digital signal processor (DSP) core for embedded applications is presented which combines the concepts of single instruction stream over multiple data streams (SIMD) and reconfigurable architec...
详细信息
ISBN:
(纸本)0769522319
A programmable parallel digital signal processor (DSP) core for embedded applications is presented which combines the concepts of single instruction stream over multiple data streams (SIMD) and reconfigurable architecture. Equipped with eight SIMD-controlled 16-bit datapaths which can also be reconfigured as two 32-bit datapaths, the DSP core can process both 16-bit and 32-bit data in parallel, showing high performance, especially in the applications preferring parallel data flow computations, such as imageprocessing. The SIMD scheme is extended with the instant-scalability of datapaths (ISSIMD), which offers the DSP a capability of dimension-controllable vector processing, so that to provide flexibility for different embedded applications. A first prototype in 0.18mum CMOS technology has been fabricated, which achieves I GMACS performance at the clock of 125MHz.
This paper presents a framework for tracking sports players in videos recorded by static domestic cameras. In the proposed approach, motion and appearance model is utilized to tracking objects as details of the object...
详细信息
ISBN:
(纸本)0889864349
This paper presents a framework for tracking sports players in videos recorded by static domestic cameras. In the proposed approach, motion and appearance model is utilized to tracking objects as details of the objects appearance is hard to obtain in low resolution videos. To assign observed objects to a trajectory, a matching process base upon the motion and appearance model is provided. In the presence of occlusion, the data association may fail to assign objects to some tracks. In this situation, the confliction is detected and some patches are generated to acquire more information to eliminate the puzzles. Then the trajectories are updated with or without observations. In the final step of tracking, new tracks are formed and some trajectories are deleted. As the foundation of tracking, Gaussian background model is utilized to segment out the players. The proposed method has been tested on real soccer game videos. The test results show that the system can successfully track multiple players independently and keep trajectories when occlusions occur.
3-D imaging LADAR systems that are capable of rapid frame acquisition may suffer from a loss of range resolution due to the duration of the pulse transmitted to the target. Because of the tradeoff between the requirem...
详细信息
ISBN:
(纸本)0819454966
3-D imaging LADAR systems that are capable of rapid frame acquisition may suffer from a loss of range resolution due to the duration of the pulse transmitted to the target. Because of the tradeoff between the requirement to produce sufficient illumination and the desire to obtain high range resolution, these systems may sacrifice range resolution in favor of improved signal to noise ratio of the detected signal. In this paper, deconvolution techniques are employed in order to obtain improved range resolution from a sequence of laser radar return images collected at extremely high speeds. The study pursued in this paper explores the degree to which range resolution can be improved in the presence of photon and speckle noise. Noise amplification in the deconvolution process serves to degrade the signal to noise ratio of the reconstructed laser radar return images. The performance of the reconstruction algorithm is quantified through the estimation of the probability of detection. It will be shown that both the probability of detection and range resolution can be improved in imaging laser radar systems.
An adaptive Neural Network system for automatic image segmentation and edge detection is proposed in this paper. This system performs image segmentation by adaptive thresholding. It automatically obtains threshold val...
详细信息
ISBN:
(纸本)088986442X
An adaptive Neural Network system for automatic image segmentation and edge detection is proposed in this paper. This system performs image segmentation by adaptive thresholding. It automatically obtains threshold values from the first order derivative of the smoothed histogram of the input image. It uses neural network architecture for adaptive thresholding. This system does not require a priori assumptions whatsoever are made about the image (type, features, contents, stochastic model, etc.). Such algorithms are most useful for applications that are supposed to work with different (and possibly initially unknown) types of images.
Classical alaorithms for identifying straight edges within an image, such as the Hough transform, run in O(n(2)) time which has been improved to O(nlog(2) (n)) by other heuristic algorithms. By focusing on 8-connected...
详细信息
ISBN:
(纸本)0889864349
Classical alaorithms for identifying straight edges within an image, such as the Hough transform, run in O(n(2)) time which has been improved to O(nlog(2) (n)) by other heuristic algorithms. By focusing on 8-connected space rather than Euclidean space, we present a method for classifying a connected list of pixels as either straight or non-straight in constant time. We then build on this method to enable a heuristic algorithm to identify the straight edges within an image in linear time. Although the improvement from O(nlog(2)(n)) to O(n) time does not appear great we find that our algorithm is faster than others even on small images due to the large amount of data required to represent an image. As images become larger or as the number of straight edges in each image increases our improvement becomes more pronounced. Our algorithm then enables fast construction of vectorial object boundaries and medial axes which, in turn, enables efficient object recognition. We illustrate this with images from RoboCup.
In this paper, we present the current status of the JPWL standardization work item. JPWL is an extension of the JPEG 2000 baseline specification in order to enable the efficient transmission of JPEG 2000 codestream ov...
详细信息
ISBN:
(纸本)0819454966
In this paper, we present the current status of the JPWL standardization work item. JPWL is an extension of the JPEG 2000 baseline specification in order to enable the efficient transmission of JPEG 2000 codestream over an error-prone network. In particular, JPWL supports a set of tools and methods for error protection and correction such as Forward Error Correcting (FEC) codes, Unequal Error Protection (UEP), and data partitioning and interleaving. We then evaluate the performance of the JPWL Error Protection Block (EPB) tool. We consider two configurations of EPB: to protect the Main and Tile-part headers, or to protect the whole codestream using UEP. Experimental results show a significant quality improvement when using EPB compared to baseline JPEG 2000.
We reported a Mexican Spanish synthesizer using concatenate diphones technique. The diphones are from an adult male with accent from the central region of Mexico. We used only 29 phonemes (we used 2 sets of vowels), a...
详细信息
ISBN:
(纸本)0889864349
We reported a Mexican Spanish synthesizer using concatenate diphones technique. The diphones are from an adult male with accent from the central region of Mexico. We used only 29 phonemes (we used 2 sets of vowels), and approximately 500 diphones. The database was design to use a small quantity of memory, to avoid the need of compressing or coding the original segments, which increases speech clarity. Grammatical rules help us design the syllable emphasis, and no special words pronunciation is needed. A simple software running under Windows was designed to write constrain-free text in a window. We have now added a PSOLA processing to improve the naturalness. The results report a natural speech, with small disadvantages compared to other more complicated international systems for Spanish, however these systems are not addressed for Mexican Spanish.
In this paper, we present an efficient compression technique that is suitable for image/video communications over wireless (mobile) channel. Our technique uses basic directional difference operators to estimate corres...
详细信息
ISBN:
(纸本)0819453617
In this paper, we present an efficient compression technique that is suitable for image/video communications over wireless (mobile) channel. Our technique uses basic directional difference operators to estimate corresponding detail subband images/videos from their approximation subband images/videos. We empirically found that the detail subband images/videos can be well approximated by the estimate subband images/videos. In this work, image and video are first decomposed using integer wavelet packet transform. Having established that detail subband images/videos can be estimated from the approximation subbands, the information needed to send over the wireless channel is only the most important subband images/videos, where we selected them via best basis selection algorithm. Next, after best basis selection, the selected subband components are encoded using either SPIHT (JPEG) for image or 3-D SPIHT for video and then the encoded data are sent over the wireless channel. The advantages of our algorithms are two folds. First, most of the computation used in our technique is performed in integer for the purpose of coding speed improvement. Second, the computation of our algorithm either SPIHT (JPEG) or 3-D SPIHT is reduced from its original computation by an order of magnitude. The reason is that in our algorithm either SPIHT (JPEG) or 3-D SPIHT is performed only on the set of important components (two or a few subband images/videos) instead of the whole image/video. Finally, we show that our proposed algorithm using SPIHT (3-D SPIHT) are better than pure JPEG (MPEG-2) both in terms of human visual image quality and computation complexity.
暂无评论