The Experts below are selected from a list of 114861 Experts worldwide ranked by ideXlab platform
F.s. Salman - One of the best experts on this subject based on the ideXlab platform.
-
Automatic Vehicle Counting from Video for Traffic Flow Analysis
2007 IEEE Intelligent Vehicles Symposium, 2007Co-Authors: M. Tekalp, F.s. SalmanAbstract:We propose a new video analysis method for counting vehicles, where we use an adaptive bounding box size to detect and track vehicles according to their Estimated Distance from the camera given the scene-camera geometry. We employ adaptive background subtraction and Kalman filtering for road/vehicle detection and tracking, respectively. Effectiveness of the proposed method for vehicle counting is demonstrated on several video recordings taken at different time periods in a day at one location in the city of Istanbul.
Viethung Dang - One of the best experts on this subject based on the ideXlab platform.
-
distributed localization in wireless sensor networks based on force vectors
International Conference on Intelligent Sensors Sensor Networks and Information Processing, 2008Co-Authors: Vietduc Le, Viethung DangAbstract:Node localization is a fundamental requirement for deploying real wireless sensor network applications such as hospital logistics, environmental monitoring, and target surveillance. However, recently proposed algorithms still could not attain enough accuracy and are too costly. To improve the previous work, we describe a low cost, high accuracy, scalable, distributed localization algorithm which based on Distance vectors. We assume that only few sensor nodes have known-locations (named as beacons) and the remaining nodes have unknown-locations (named as normal nodes). Each node updates its location from currently Estimated Distance vectors and given pairwise Distances which are derived from received signal strength (RSS) measurements or time of arrival (TOA), and then passes this new location to neighbors until achieving enough convergence. Analysis, simulation and experimental results show that the proposed algorithm outperforms the other range-based algorithms. Specially the proposed algorithm on real-world experiment measurements, which contain unpredictable noise and shadowing, achieves better accuracy than the previous work do. The proposed method can perform well even with only few reference devices or anchors.
M. Tekalp - One of the best experts on this subject based on the ideXlab platform.
-
Automatic Vehicle Counting from Video for Traffic Flow Analysis
2007 IEEE Intelligent Vehicles Symposium, 2007Co-Authors: M. Tekalp, F.s. SalmanAbstract:We propose a new video analysis method for counting vehicles, where we use an adaptive bounding box size to detect and track vehicles according to their Estimated Distance from the camera given the scene-camera geometry. We employ adaptive background subtraction and Kalman filtering for road/vehicle detection and tracking, respectively. Effectiveness of the proposed method for vehicle counting is demonstrated on several video recordings taken at different time periods in a day at one location in the city of Istanbul.
Vietduc Le - One of the best experts on this subject based on the ideXlab platform.
-
distributed localization in wireless sensor networks based on force vectors
International Conference on Intelligent Sensors Sensor Networks and Information Processing, 2008Co-Authors: Vietduc Le, Viethung DangAbstract:Node localization is a fundamental requirement for deploying real wireless sensor network applications such as hospital logistics, environmental monitoring, and target surveillance. However, recently proposed algorithms still could not attain enough accuracy and are too costly. To improve the previous work, we describe a low cost, high accuracy, scalable, distributed localization algorithm which based on Distance vectors. We assume that only few sensor nodes have known-locations (named as beacons) and the remaining nodes have unknown-locations (named as normal nodes). Each node updates its location from currently Estimated Distance vectors and given pairwise Distances which are derived from received signal strength (RSS) measurements or time of arrival (TOA), and then passes this new location to neighbors until achieving enough convergence. Analysis, simulation and experimental results show that the proposed algorithm outperforms the other range-based algorithms. Specially the proposed algorithm on real-world experiment measurements, which contain unpredictable noise and shadowing, achieves better accuracy than the previous work do. The proposed method can perform well even with only few reference devices or anchors.
Mrinalini Balki - One of the best experts on this subject based on the ideXlab platform.
-
ultrasound imaging of the thoracic spine in paramedian sagittal oblique plane the correlation between Estimated and actual depth to the epidural space
Regional Anesthesia and Pain Medicine, 2011Co-Authors: Aliya Salman, Cristian Arzola, Uma Tharmaratnam, Mrinalini BalkiAbstract:Background: Ultrasound (US) imaging of the spine has been shown to be a reliable tool to facilitate lumbar epidural needle placement; however, its feasibility in thoracic epidural placement is still unknown. The objective of this study was to assess the accuracy and reliability of prepuncture US imaging in the paramedian sagittal oblique plane to estimate the depth to the epidural space and optimum insertion point for guiding epidural needle placement at the mid-low thoracic level. Methods: This prospective study included 35 healthy adult patients who requested thoracic epidural analgesia before their upper abdominal surgeries. Ultrasound imaging was done in the paramedian sagittal oblique plane at the desired thoracic level to identify the intervertebral space, the Distance from the skin to the epidural space (US depth [UD]) and the needle insertion point. Subsequently, a staff anesthesiologist located the epidural space through the predetermined insertion point and marked the actual Distance from the skin to the epidural space (needle depth [ND]) on the needle with a sterile marker. The agreement between the UD and the ND was calculated using the Pearson and concordance correlation coefficients and Bland-Altman analysis with 95% limits of agreement. Results: The average patient age was 56 (SD, 14) years, and body mass index was 28 (SD, 6) kg/m2. The precision of the agreement between UD and ND Estimated by Pearson correlation coefficient was 0.75, and the accuracy was 0.80, whereas the concordance correlation coefficient was 0.60 (confidence interval, 0.43-0.78). The mean UD and ND were 4.3 (SD, 0.96) and 5.0 (SD, 1.2) cm, respectively. The Bland-Altman analysis showed a mean difference of −0.71 cm (95% limits of agreement, 0.8 to −2.2 cm). There was a significant direct correlation of the ND with the body mass index (r2 = 0.27, P = 0.008). The mean number of attempts was 1 (p25-p75 = 1-2), and the epidural space was identified with 2 or less redirections in 88% of the cases. Conclusions: We found a good correlation between the US-Estimated Distance to the epidural space and the actual measured needle Distance in our patients. We suggest that our proposed prepuncture US method, using the paramedian sagittal oblique approach, can be a useful guide to facilitate the placement of epidural needles at mid-low thoracic levels. A randomized controlled trial is necessary to confirm the utility of prepuncture US in thoracic epidural placement.
-
ultrasound imaging of the lumbar spine in the transverse plane the correlation between Estimated and actual depth to the epidural space in obese parturients
Annual Meeting of the Society for Obstetric Anesthesia and Perinatology, 2009Co-Authors: Mrinalini Balki, Yung Lee, Stephen H Halpern, Jose C A CarvalhoAbstract:BACKGROUND: Prepuncture lumbar ultrasound scanning is a reliable tool to facilitate labor epidural needle placement in nonobese parturients. In this study, we assessed prepuncture lumbar ultrasound scanning as a tool for estimating the depth to the epidural space and determining the optimal insertion point in obese parturients. METHODS: We studied 46 obese parturients, with prepregnancy body mass index (BMI) >30 kg/m 2 , requesting labor epidural analgesia. Ultrasound imaging was done by one of the investigators to identify the midline, the intervertebral space, and the Distance from the skin to the epidural space (ultrasound depth, UD) at the level of L3-4. Subsequently, an anesthesiologist blinded to the UD located the epidural space through the predetermined insertion point and marked the actual Distance from the skin to the epidural space (needle depth, ND) on the needle with a sterile marker. The agreement between the UD and the ND was calculated using the Pearson correlation coefficient and a paired t-test. Bland-Altman analysis was used to determine the 95% limits of agreement between the UD and the ND. RESULTS: The prepregnancy BMI ranged from 30 to 79 kg/m 2 , and the BMI at delivery was 33-86 kg/m 2 . The Pearson correlation coefficient between the UD and the ND was 0.85 (95% confidence interval: 0.75-0.91), and the concordance correlation coefficient was 0.79 (95% confidence interval: 0.71-0.88). The mean (±SD) ND and UD were 6.6 ± 1.0 cm and 6.3 ± 0.8 cm, respectively (difference = 0.3 cm, P = 0.002). The 95% limits of agreement were 1.3 cm to ―0.7 cm. Epidural needle placement using the predetermined insertion point was done without reinsertion at a different puncture site in 76.1% of parturients and without redirection in 67.4%. CONCLUSIONS: We found a strong correlation between the ultrasound-Estimated Distance to the epidural space and the actual measured needle Distance in obese parturients. We suggest that prepuncture lumbar ultrasound may be a useful guide to facilitate the placement of epidural needles in obese parturients.