Home OALib Journal OALib PrePrints Submit Ranking News My Lib FAQ About Us Follow Us+
 Title Keywords Abstract Author All
Search Results: 1 - 10 of 100 matches for " "
 Page 1 /100 Display every page 5 10 20 Item
 计算机科学 , 2009, Abstract: This paper presented a new algorithm to estimate the standard deviation of noise for nature images.It uses a gradient filter to estimate the noise on some parts which are selected by a controlled function from the tessellating ima-ge.The controlled function can be related to the preestimate of SNR.We optimally determined the controlled function as an exponential function by minimizing the total error.The controlled function can effectively offset the effect which is caused by filtering the image structure.T...
 Computer Science , 2015, Abstract: Recent studies have shown that Convolutional Neural Networks (CNNs) are vulnerable to a small perturbation of input called "adversarial examples". In this work, we propose a new feedforward CNN that improves robustness in the presence of adversarial noise. Our model uses stochastic additive noise added to the input image and to the CNN models. The proposed model operates in conjunction with a CNN trained with standard backpropagation algorithm. In particular, convolution, max-pooling, and ReLU layers are modified to benefit from the noise model. Our model is parameterized by only a mean and variance per pixel which simplifies computations and makes our method scalable to a deep architecture. The proposed model outperforms the standard CNN by 13.12% on ImageNet and 7.37% on CIFAR-10 under adversarial noise at the expense of 0.28% of accuracy drop when used in the original dataset -- with no added noise.
 Computer Science , 2015, Abstract: Deep networks have been successfully applied to visual tracking by learning a generic representation offline from numerous training images. However the offline training is time-consuming and the learned generic representation may be less discriminative for tracking specific objects. In this paper we present that, even without offline training with a large amount of auxiliary data, simple two-layer convolutional networks can be powerful enough to develop a robust representation for visual tracking. In the first frame, we employ the k-means algorithm to extract a set of normalized patches from the target region as fixed filters, which integrate a series of adaptive contextual filters surrounding the target to define a set of feature maps in the subsequent frames. These maps measure similarities between each filter and the useful local intensity patterns across the target, thereby encoding its local structural information. Furthermore, all the maps form together a global representation, which is built on mid-level features, thereby remaining close to image-level information, and hence the inner geometric layout of the target is also well preserved. A simple soft shrinkage method with an adaptive threshold is employed to de-noise the global representation, resulting in a robust sparse representation. The representation is updated via a simple and effective online strategy, allowing it to robustly adapt to target appearance variations. Our convolution networks have surprisingly lightweight structure, yet perform favorably against several state-of-the-art methods on the CVPR2013 tracking benchmark dataset with 50 challenging videos.
 Physics , 2013, DOI: 10.1103/PhysRevLett.111.093604 Abstract: Interactions between atoms and lasers provide the potential for unprecedented control of quantum states. Fulfilling this potential requires detailed knowledge of frequency noise in optical oscillators with state-of-the-art stability. We demonstrate a technique that precisely measures the noise spectrum of an ultrastable laser using optical lattice-trapped $^{87}$Sr atoms as a quantum projection noise-limited reference. We determine the laser noise spectrum from near DC to 100 Hz via the measured fluctuations in atomic excitation, guided by a simple and robust theory model. The noise spectrum yields a 26(4) mHz linewidth at a central frequency of 429 THz, corresponding to an optical quality factor of $1.6\times10^{16}$. This approach improves upon optical heterodyne beats between two similar laser systems by providing information unique to a single laser, and complements the traditionally used Allan deviation which evaluates laser performance at relatively long time scales. We use this technique to verify the reduction of resonant noise in our ultrastable laser via feedback from an optical heterodyne beat. Finally, we show that knowledge of our laser's spectrum allows us to accurately predict the laser-limited stability for optical atomic clocks.
 Mathematics , 2007, DOI: 10.1016/j.spa.2008.01.005 Abstract: We obtain several extensions of Talagrand's lower bound for the small deviation probability using metric entropy. For Gaussian processes, our investigations are focused on processes with sub-polynomial and, respectively, exponential behaviour of covering numbers. The corresponding results are also proved for non-Gaussian symmetric stable processes, both for the cases of critically small and critically large entropy. The results extensively use the classical chaining technique; at the same time they are meant to explore the limits of this method.
 Computer Science , 2012, Abstract: The most median-based de noising methods works fine for restoring the images corrupted by Randomn Valued Impulse Noise with low noise level but very poor with highly corrupted images. In this paper a directional weighted minimum deviation (DWMD) based filter has been proposed for removal of high random valued impulse noise (RVIN). The proposed approach based on Standard Deviation (SD) works in two phases. The first phase detects the contaminated pixels by differencing between the test pixel and its neighbor pixels aligned with four main directions. The second phase filters only those pixels keeping others intact. The filtering scheme is based on minimum standard deviation of the four directional pixels. Extensive simulations show that the proposed filter not only provide better performance of de noising RVIN but can preserve more details features even thin lines or dots. This technique shows better performance in terms of PSNR, Image Fidelity and Computational Cost compared to the existing filters.
 Physics , 2009, DOI: 10.1103/PhysRevA.81.033826 Abstract: We analyze the effects of noise and parameter deviations in a bichromatic Raman type white light cavity, with potential applications in precision measurements such as gravitational wave detection. The results show that the dispersion variation induced by parameter deviation can be controlled within $10^{-4}$. The laser phase noise decreases the dispersion magnitude while the amplitude noise increases it. Although we can always adjust the parameters to satisfy the white light condition, both noises make the cavity transmission curve uneven.
 Advances in Sexual Medicine (ASM) , 2013, DOI: 10.4236/asm.2013.31004 Abstract: Objectives: We propose a surgical correction of the penile curvature applying a technique based on the 16-dot plication technique modified by burying the knots in a shallow trough of incised tunica. We entitle this modification the “Kiel Knots”. Material and methods: 20 patients with a penile deviation, average age 36.8 years (24 - 52) were operated. Follow-up time was 26 months. In 8 patients the deviation was congenital, in 12 patients it was an acquired deviation. The deviation was >30? in all patients. Surgical technique: A circumcising incision was made and the penis was degloved. Buck’s fascia was incised exposing the tunica albuginea opposite the curvature. 8 dots were marked bilaterally on the tunica albugineaand a5 mmincision along the marks was made without cutting into the Corpus cavernosum. Instead of plicating with one suture for 4 dots, our modification uses one suture for two dots with the knot buried in a shallow trough created by a scalpel. Results: The average operation time was 64 minutes. We observed a loss of penile length in 30% of the patients (0.5 - 1 cm). There were no problems with erectile function. In a follow-up of 2 years, 90% of the patients remained without recurrence of deviation. None of patients reported problems with the suture knots. Conclusion: Our technique achieves penile straightening with minimal loss of length and no erectile dysfunction.
 International braz j urol , 2003, DOI: 10.1590/S1677-55382003000400009 Abstract: purpose: the schroeder-essed plication procedure is a standard technique for the correction of penile curvature. in a retrospective analysis we compared functional results and quality of life (lq) of the original technique with inverted sutures as described by schroeder-essed and our slight modification consisting of horizontal incisions into the tunica albuginea. materials and methods: twenty-six patients with congenital penis deviation were treated for penile deviation by the original schroeder-essed plication with inverted sutures (11 patients) and by the described modification (15 patients). in case of modified technique, horizontal and parallel incisions 4 mm to 6 mm apart and about 8 mm - 10 mm long were made through the tunica albuginea. the outer edges of the incisions were then approximated with permanent inverted sutures (gore-tex？ 3-0). mean age was 21.6 years in the first group and 23.2 years in the second group. average follow-up was 28 months and 13 months, respectively. the preoperative penile deviation angle was > 25o in all patients without difference between the 2 groups. results: all patients in both groups reported an improvement in their quality of life and full ability to engage in sexual intercourse. nine patients (88%) in the first group and 14 patients (93%) in the second group were satisfied with the cosmetic result. in contrast, 10 patients (91%) of the first and 13 patients (87%) of the second group complained of penile shorting. recurrence of deviation was only noticed in 2 males in the first group (18%). conclusions: our results indicate that this simple modification of the schroeder-essed plication offers good functional and cosmetic results. most patients were satisfied with the penile angle correction results.
 International braz j urol , 2003, Abstract: PURPOSE: The Schroeder-Essed plication procedure is a standard technique for the correction of penile curvature. In a retrospective analysis we compared functional results and quality of life (LQ) of the original technique with inverted sutures as described by Schroeder-Essed and our slight modification consisting of horizontal incisions into the tunica albuginea. MATERIALS AND METHODS: Twenty-six patients with congenital penis deviation were treated for penile deviation by the original Schroeder-Essed plication with inverted sutures (11 patients) and by the described modification (15 patients). In case of modified technique, horizontal and parallel incisions 4 mm to 6 mm apart and about 8 mm - 10 mm long were made through the tunica albuginea. The outer edges of the incisions were then approximated with permanent inverted sutures (Gore-Tex 3-0). Mean age was 21.6 years in the first group and 23.2 years in the second group. Average follow-up was 28 months and 13 months, respectively. The preoperative penile deviation angle was > 25o in all patients without difference between the 2 groups. RESULTS: All patients in both groups reported an improvement in their quality of life and full ability to engage in sexual intercourse. Nine patients (88%) in the first group and 14 patients (93%) in the second group were satisfied with the cosmetic result. In contrast, 10 patients (91%) of the first and 13 patients (87%) of the second group complained of penile shorting. Recurrence of deviation was only noticed in 2 males in the first group (18%). CONCLUSIONS: Our results indicate that this simple modification of the Schroeder-Essed plication offers good functional and cosmetic results. Most patients were satisfied with the penile angle correction results.
 Page 1 /100 Display every page 5 10 20 Item