توجه: محتویات این صفحه به صورت خودکار پردازش شده و مقاله‌های نویسندگانی با تشابه اسمی، همگی در بخش یکسان نمایش داده می‌شوند.
۱Adaptive Rate Allocationfor Unequal Budget Images Transmissionover Time–Varying BSC Channels
نویسنده(ها): ،
اطلاعات انتشار: هفتمین کنفرانس بین المللی پیشرفتهای علوم و تکنولوژی، سال
تعداد صفحات: ۸
An efficient rate allocation algorithm for the progressive transmission of multiple images with unequal budgets over time–varying BSC channels is proposed. The algorithm is linear–time in the number of transmitted packets per image and its rate allocation solution for each image can achieve a performance equal or very close to the distortion optimal solution for that image. Our simulations for the transmission of images, encoded by embedded source coders, over the binary symmetric channel (BSC) show that with very low complexity the proposed algorithm successfully adapts the channel code rates to the changes of the channel parameter<\div>

۲Noise Reduction and Kidney Ultrasound Image Enhancement Based on Multi–Scale Transform and Curvelet Transform
نویسنده(ها): ،
اطلاعات انتشار: هفتمین کنفرانس بین المللی پیشرفتهای علوم و تکنولوژی، سال
تعداد صفحات: ۸
A new algorithm for contrast enhancement and noise removal from the ultrasound images is proposed. At first, Sticks filteris used to remove the initial noise from the ultrasound images. Then the image brightness is improved. Finally multi–scale and new generation of curvelet transform, the coefficient adaptive correction of these transforms, non–linear modificationand adaptive functions suggested here are applied to enhance the final contrast and removing any noise from the kidney ultrasound images.<\div>

۳Improving the Performance of Q–learning Using Simultanouse Q–values Updating
اطلاعات انتشار: دومین کنفرانس بین المللی شبکه های اطلاعاتی هوشمند و سیستم های پیچیده، سال
تعداد صفحات: ۶
Q–learning is a one of the best model–free reinforcement learning algorithms. The goal is to find an estimate of the optimal action–value function called Q–value function. The Q–value function is defined as the expected sum of future rewards obtained by taking an action in the current state. The main drawback of Q–learning is that the learning process is expensive for the agent, specially, in the beginning steps. Because, every state–action pair should be visited frequently in order to converge to the optimal policy. In this paper, the concept of opposite action is used to improve the performance of the Q–learning algorithm, especially, in the beginning steps of the learning. Opposite actions suggest updating two Q–values, simultaneously. The agent will update Q–value for each action and corresponding opposite action and thus increasing the speed of learning. The novel Q–learning method based on the concept of opposite action is simulated for the famous test–bed grid world problem. The results show the ability of the proposed method to improve the learning process<\div>

۴Designing the Fuzzy Rule Base Using LAFA
اطلاعات انتشار: کنفرانس بین المللی علوم مهندسی، هنر و حقوق، سال
تعداد صفحات: ۶
Having inaccurate and faulty knowledge and being capable of answering vague and ambiguous queries in many fields such as pattern recognition, automatic control and decision analysis, fuzzy databases are usually faced with a series of uncertainties in parameters, system structure and the environment in which the system operates. Many different designs have been proposed for databases. Among all of them, the most difficult action is to design theoptimal fuzzy rule base. Using Learning Automata Firefly Algorithm (LAFA), the research aimed to design a fuzzy rule base in a fuzzy controller of TSK type. The performance and efficiency of the proposed method were obtained through implementing it on the benchmarkand comparing the results with those of other methods (taken from different papers). They indicated a smaller number of fuzzy rules and decreased error control<\div>

۵A Survey of Feature Extraction Techniques in OCR
نویسنده(ها): ، ،
اطلاعات انتشار: اولین کنفرانس بین المللی دستاوردهای نوین پژوهشی در مهندسی برق و کامپیوتر، سال
تعداد صفحات: ۶
Over the last five years optical character recognition approaches have been under gone an enormous number of changes. Many efforts have been done and a wide range of algorithms have been used in order to improve the performance of existing methods of OCR in many languages. This paper presents an overview of feature extraction methods for character recognition in different texts. The feature extraction stage is an important component of any recognition system. It is also very much dependent on the task, input, and recognition algorithm used. The feature extraction methods are discussed in terms of invariance properties and expected distortions and variability of the characters.<\div>
نمایش نتایج ۱ تا ۵ از میان ۵ نتیجه