In this paper, an improved wavelet based medical image watermarking algorithm is proposed. Initially, the proposed technique decomposes the cover medical image into ROI and NROI regions and embedding three different watermarks into the Non-Region Of Interest (NROI) part of the transformed DWT cover image for compact and secure medical data transmission in E-health environment. In addition, the method addressing the problem of channel noise distortion may lead to faulty watermark by applying Error Correcting Codes (ECCs) before embedding them into the cover image. Further, the Bit Error Rates (BER) performance of the proposed method is determined for different kind of attacks including ‘Checkmark’ attacks. Experimental results indicate that the Turbo code performs better than BCH (Bose- Chaudhuri-Hochquenghem) error correction code. Furthermore, the experimental results validate the effectiveness of the proposed framework in terms of BER and embedding.
Video scalability is a recent video coding technology that allows content providers to oﬀer multiple quality versions from a single encoded video file in order to target diﬀerent kinds of end-user devices and networks. One form of scalability utilizes the region- of-interest concept, that is, the possibility to mark objects or zones within the video as more important than the surrounding area. The scalable video coder ensures that these regions-of-interest are received by an end-user device before the surrounding area and preferably in higher quality. In this paper, novel algorithms are presented making it possible to automatically track the marked objects in the regions of interest. Our methods detect the overall motion of a designated object by retrieving the motion vectors calculated during the motion estimation step of the video encoder. Using this knowledge, the region-of-interest is translated, thus following the objects within. Furthermore, the proposed algorithms allow adequate resizing of the region-of-interest. By using the available information from the video encoder, object tracking can be done in the compressed domain and is suitable for real-time and streaming applications. A time-complexity analysis is given for the algorithms proving the low complexity thereof and the usability for real-time applications. The proposed object tracking methods are generic and can be applied to any codec that cal- culates the motion vector field. In this paper, the algorithms are implemented within MPEG-4 fine-granularity scalability codec. Diﬀerent tests on diﬀerent video sequences are performed to evaluate the accuracy of the methods. Our novel algorithms achieve a precision up to 96.4%.
Our technique is intended to capture Associate in Nursing exploit the temporal flow of events associated with the abandonment of an object. within the planned algorithmic rule, initial Region of interest is chosen, then video is segmental into frames, shots or pictures for process background subtraction is performed to cite any new object which will have entered the scene. Afterward the determination or extraction of foreground image objects are half- tracked by blob analysis and at last abandoned objects are detected.
We have implemented a new improved WDR Algorithm with Region of interest concept on rice image to calculate CR, MSE, BPP, and PSNR. We have used MATLAB 2011 wavelet toolbox for implementing these methods. Figure 2 shows an original Natural image compressed by WDR Algorithm Figure 3 shows an original Natural image and compressed image using our proposed (WDR-ROI) Algorithm.
Abstract: Medical images do contain important and unimportant spatial regions. Compression methods which are capable of reconstructing the image with high quality are required to compress the medical images. For these images, only a portion of it is useful for diagnosis hence a region based coding techniques are significant for compressing and transmission. Extracting a significant region is of great demand since a slighter mistake may leads to wrong diagnosis. This paper is focused on investigating multiple image processing algorithms for medical images. All the images may not contain the same region of interest, so different approaches are supposed to apply for different images. In this three types of medical images were considered like magnetic resonance (MR) brain images, computer tomography (CT) abdomen images and X-ray lung images. In this paper three automatic region of interest extraction algorithms were proposed for different types of images.
A region of interest (often abbreviated ROI), could be a designated set of samples at intervals a dataset known for a specific purpose. The concept of associate degree ROI is often employed in several application areas. as an example, in medical imaging, the boundaries of a tumour is also outlined on a picture or in a very volume, for the aim of measurement its size. The endocardial border is also outlined on a picture, maybe throughout totally different phases of the oscillation, as an example end-systole and end- diastole, for the aim of assessing viscus perform. In geographical data systems (GIS), associate degree ROI will be taken virtually as a two-dimensional figure choice from a second map. Previous works for ROI detection in remote sensing pictures square measure inaccurate and prohibitively computationally advanced. thence we tend to propose region-of-interest extraction technique supported frequency domain analysis and salient region detection (FDA-SRD) technique for ROI extraction. For this, the photographs square measure regenerate from RGB to HIS as preprocessing. The prominence driven image ensuing scale area generally preserves or perhaps enhances semantically vital structures like edges, lines, or flow-like structures within the foreground, and inhibits and smoothest muddle within the background. The image is reconstructed victimisation fusion supported the initial image, the image at the ultimate scale at that the diffusion method converges, and therefore the image at a midscale. Our algorithmic program emphasizes the foreground options, that square measure vital for image classification. The background image regions, whether or not thought-about as contexts
This paper classified the videos into one of four types based on region of interest: central focus, peripheral focus, lower focus, and upper focus. Then, this paper allocated the quantization step-size (parameter) to the region of interest differentially without an abrupt change between the ROI and adjacent regions. Also, this paper demonstrated an improvement in subjective picture quality by adapting the step-size, compared to the existing method of allocating the quantization parameter. This paper verified that a reasonable difference between the maximum and the minimum values of the quantization parameter within a video was about 4-8 in the case of a differentially applied quantization parameter. Although this paper focused on the H.264 video coding, it can be directly used to other types of video coding standard. In addition, it can be expected that this paper will be extensively applied with other ROI extraction methods and the macroblock-based control besides the slice-based control.
In this study, excellent agreement was found between two common methods used to define the regions of interest of the multifidus and erector spinae muscle from axial MRIs. Inclusion of the fat in-between the epi- myseal border and the fascial plane results in larger values for tCSA, FSF, fCSA and mCSA when compared to excluding the area of fat, with no differences in vari- ance. The decision to include or exclude the fat area from a region of interest measurement of the lumbar muscles should be made based on the primary outcome a researcher is interested in measuring. Inclusion of the fat area results in a more gross measure of fatty
Abstract—The increasing availability of digital pathology im- ages has motivated the design of tools to foster multidisciplinary collaboration among researchers, pathologists, and computer scientists. Telepathology plays an important role in the develop- ment of collaborative tools, as it facilities the transmission and access of pathology images by multiple users. However, the huge ?le size associated with pathology images usually prevents full exploitation of the collaborative telepathology system potential. Within this context, rate control (RC) is an important tool that allows meeting storage and bandwidth requirements by controlling the bit rate of the coded image. In this paper, we propose a novel graph-based RC algorithm with lossless region of interest (RoI) coding for pathology images. The algorithm, which is designed for block-based predictive transform coding methods, compresses the non-RoI in a lossy manner according to a target bit rate and the RoI in a lossless manner. It employs a graph where each node represents a constituent block of the image to be coded. By incorporating information about the coding cost similarities of blocks into the graph, a graph kernel is used to distribute a target bit budget among the non-RoI blocks. In order to increase RC accuracy, the algorithm uses a rate-lambda (R-λ) model to approximate the slope of the rate-distortion curve of the non-RoI, and a graph-based approach to guarantee that the target bit rate is accurately attained. The algorithm is implemented in the HEVC standard and tested over a wide range of pathology images with multiple RoIs. Evaluation results show that it outperforms other state-of-the-art methods designed for single images by very accurately attaining the target bit rate of the non-RoI.
quantitatively [1 – 20]. Because of the heterogeneity of tumor tissues, ADC measurements may depend on region of interest (ROI) selection. In DWI, ROIs ob- tained by three main ROI techniques, i.e. whole-volume, single-slice, and small solid-sample, have been applied for obtaining the ADC values of tumors [21, 22]. However, it is rarely assessed for thyroid nodules. The current study aimed to evaluate the effect of ROI selec- tion on ADC measurements and interobserver variability in thyroid nodules.
The region where the proﬁle of a horn possesses better RF performance can be termed as the ‘Region of Interest (RoI)’ of a horn. Based on simulated results of the above mentioned proﬁles, it has been observed that the performance of a horn depends on two of the proﬁling sections (i) throat section and (ii) aperture section. After a rigorous study, the regions of interest for both these sections are obtained with a target to achieve the best gain and side-lobe performance.
When an image or scene is viewed, the eye gaze tends to pause on small regions within the image, called fixation areas. On average, fixations last for around 200 ms during the reading of linguistic text, and 350 ms during the viewing of a scene . Existing approaches for detecting Region of Interest (ROI) in the viewed image first represent the centre of a fixation area as a fixation point , and then use clustering to group these fixation points from all fixation areas into spatial regions, identified as ROI. Various clustering approaches have been used to detect ROI, such as k-means and distance threshold , , Density-based spatial clustering of applications with noise (DB-SCAN) , Distance-Threshold Identification (I- DT)  and Mean-shift . The gaze data, acquired by commercial eye trackers, is normally affected by high level of measurement noise and contains missing data due to eye blinks and occasional head movements. This motivates the use of Graph Signal Processing (GSP), an emerging field used to represent irregular data structures on graphs , , for robust gaze data clustering.
This paper presents a video bit-rate transcoder for baseline profile in H.264/AVC standard to fit the available channel bandwidth for the client when transmitting video bit-streams via communication channels. To maintain visual quality for low bit-rate video efficiently, this study analyzes the decoded information in the transcoder and proposes a Bayesian theorem-based region-of-interest (ROI) determination algorithm. In addition, a curve fitting scheme is employed to find the models of video bit-rate conversion. The transcoded video will conform to the target bit-rate by re-quantization according to our proposed models. After integrating the ROI detection method and the bit-rate transcoding models, the ROI-based transcoder allocates more coding bits to ROI regions and reduces the complexity of the re-encoding procedure for non-ROI regions. Hence, it not only keeps the coding quality but improves the efficiency of the video transcoding for low target bit-rates and makes the real-time transcoding more practical. Experimental results show that the proposed framework gets significantly better visual quality.
Conﬁguration of problem geometry is illustrated as in Figure 1. Lossy and homogeneous objects are considered for analyses under an ideal situation in which scattered ﬁelds are not aﬀected by noise. In this research framework, lung area is considered as the original reconstruction region, and a tumour of 8 mm in radius is the object to seek. Both lung and tumour are embedded in a free space medium to allow perfect signal propagation. The lung is surrounded by M = 12 point source antennas. Each point source antenna is positioned with the same distance at r t m (m = 12, . . . , M) in a radial formation. Only one antenna can transmit modulated Gaussian pulse signal based impressed current formulated as in Eq. (1) towards the region of interest (ROI) at a time.
While the methods mentioned previously achieve excellent results on semantic segmentation, all of them rely on pre-trained data to achieve the required results. As a consequence, they have a large GPU RAM requirement because they depend on adding layers to models that were trained on classification tasks. We are interested in developing region of interest (ROI) model that have lower memory requirement. These models will be used as a means of pre- processing the data to simplify the classification problem (in this thesis: the whale recognition problem (Chapters 2 and 3) and the left ventricle volume estimation problem (Chapters 4 and 5)). Therefore, these models need to deliver the results as fast as possible and with low memory consumption. In other words, detecting the region of interest is a sub-task that can simplify the main problem (Right whale recognition or left ventricle estimation) and we are interested in an e ffi cient solution that can perform this task.
Abstract: - Ultrasound scanning of the kidney is performed to assess kidney size, shape and location as well as to detect any abnormalities in kidney like cysts and stones. Since ultrasound image contains speckle noise, performing the segmentation methods for the kidney images has always been a very challenging task. For further segmentation purpose, deleting and removing the complicated background not only speeds up the segmentation process, but also increases accuracy. However, in previous studies, the ROI of the kidney is manually cropped. Therefore, this study proposed an automatic region of interest (ROI) generation for kidney ultrasound images. The methods consist of the speckle noise reduction using Gaussian low-pass filter, texture analysis by calculating the local entropy of the image, threshold selection, morphological operations, object windowing, determination of seed point and last but not least the ROI generation. This algorithm has been tested to more than 200 kidney ultrasound images. As the result, for longitudinal kidney images, out of 120 images, 109 images generate true ROI (91%) and another 11 images generate false ROI (9%). For transverse kidney images, out of 100 images, 89 images generate true ROI (89%) and 11 images generate false ROI (11%). To conclude, the method in this study can be practically used for automatic generation of US kidney ROI.
Shirani also presented a non-linear PSS-MDC method which investigated its performance by evaluating the case where there were one or more missing descriptions . According to his work, some parts of a frame which are more important called region of interest (ROI) were sampled with a greater rate (based on an exponential equation) compared to other parts of the frame. In other words, descriptions include more information regarding the ROI parts of the frame resulting in an enhancement of the side quality. More importantly, this method provides for greater performance with regards to the subjective assessment by the human eye since objects and not pixels are more emphasized. Although Shirani’s method pro- vides for the enhancement of the side quality, he did not discuss how the ROI parts of a frame were detected which is important when involving fast video contents or live video streaming. In this paper, we provide a new spatial MDC algorithm that adds redundancy to the descriptions more practically for 3D videos.
Methods: Thirty healthy volunteers (mean age 37.8 years, SD 11.4) underwent DTI of the brain with 3T MRI. Region-of-interest (ROI) -based measurements were calculated at eleven anatomical locations in the pyramidal tracts, corpus callosum and frontobasal area. Two ROI-based methods, the circular method (CM) and the freehand method (FM), were compared. Both methods were also compared by performing measurements on a DTI phantom. The intra- and inter-observer variability (coefficient of variation, or CV%) and repeatability (intra-class correlation coefficient, or ICC) were assessed for FA and ADC values obtained using both ROI methods. Results: The mean FA values for all of the regions were 0.663 with the CM and 0.621 with the FM. For both methods, the FA was highest in the splenium of the corpus callosum. The mean ADC value was 0.727 ×10 -3 mm 2 /s with the CM and 0.747 ×10 -3 mm 2 /s with the FM, and both methods found the ADC to be lowest in the corona radiata. The CV percentages of the derived measures were < 13% with the CM and < 10% with the FM. In most of the regions, the ICCs were excellent or moderate for both methods. With the CM, the highest ICC for FA was in the posterior limb of the internal capsule (0.90), and with the FM, it was in the corona radiata (0.86). For ADC, the highest ICC was found in the genu of the corpus callosum (0.93) with the CM and in the uncinate fasciculus (0.92) with FM.
using the same images (no rescan), individual region of interest place- ment was carried out by one investigator 4 –7 days after the first as- sessment, and average DTI parameters were calculated for each region of interest. The 2 region of interest placements were conducted inde- pendently from each other (ie, information concerning the first place- ment was not accessible at the second time, and the investigator was blinded regarding all other clinical data). To calculate interobserver variability, region of interest placement and DTI parameter compu- tation was accomplished by a second independent investigator. MR Imaging Data Acquisition
The motion estimation stage seeks to obtain an educated guess of the direction of the object’s motion. To do this, it is used a square section called region of interest (ROI) to represent the object being tracked. Additionally, the ROI, which is placed in the object being tracked in the current frame, is also located in the same position but in next frame. Then, it is analyzed the changing pixels in the ROIs to estimate the direction of the object’s motion. As a way of explaining the proposed algorithm, it is shown in the Figure 1(a) the current frame of a virtual video where the object to be tracked is represented by a region of interest. In the same way, the Figure 1(b)