انجمن کامپیوتر ایران

مبحث ذخيره سازي داده امروزه به مسأله بسيار مهمي در مراكز داده اينترنتي و سيستم هاي محاسبات فو ق سريع تبديل شده است. چرا كه هرساله حجم داده هاي ذخيره شده در دنيا تقريبًٌا دو برابر مي شود و در يك سيستم محاسباتي ممكن است متجاوز از يك پتابايت ١ باشد. از جمل ه ي چنين سيست م هايي در سازما ن هاي هواشناسي و پايگاه هاي نظامي ممكن است يافت شوند. اين مقاله روشي جديد ارائه مي كند كه بر اساس آن در يك سيستم محاسباتي كلاستري، زمان مورد نياز براي نقل و انتقالات داده در مجموع كاهش م يياب د. روش پيشنهادي جديد عبارت است از مهاجرت پردازش هاي درحال اجرا بر روي گره هاي پردازشي با استفاده از فن آوري جديد مجازي سازي كه به منظور توازن بار در شبكه استفاده مي شود. نتايج حاصله از شبب ه سازي نشان مي دهد كه اين روش جديد كاهش قابل ملاحظه اي در زمان مورد نياز سيستم محاسبات ي جهت انجام پردازش داده در مقاط ع زماني مشخص ايجاد مي كند.

نكته حايز اهميت در تشخيص خودرو استفاده از تكنيكهاي پايدار در برابر تغييرات است . در اين نوشتار، به منظور تشخيص خودروها از ويژگي تقارن استفاده شده كه اين ويژگي در برابر تغييرات نور پايدار است. همچنين روشي موثر براي برخورد با مشكل همپوشاني خودروها است. براي يافتن خط تقارن خودروها از روش تشخيص سريع خط تقارن استفاده نمودهايم. اين روش به كمك تبديل هاف و استفاده از پيكسلهاي لبه، خط تقارن هر شي را بدست مي آورد. همچنين الگوريتم بهبود تشخيص سريع خط تقارن به منظور رفع خطاهاي ناشي از عمق تصوير ارايه گرديده است كه الگوريتم تشخيص سريع خط تقارن قادر به رفع آن نبوده است. به منظور رديابي خودروها از فيلتر كالمن استفاده شده است. آزمايشات انجام شده نشان داده است كه روش پيشنهادي قادر است با دقت 96.52 % تعداد زيادي خودرو را در يك جاده رديابي كند.

در اين مقاله يك مرز فعال پارامتري جديد بنام مرز فعال خودنسبي معرفي و تكميل گرديده است. در اين روش با تركيب سامانه نگاشت خودنسبي و تبديل موجك، يك ميدان نيروي خارجي براي مدل مارها ارائه شده است. همچنين، حجم محاسباتي روش پيشنهادي پس از پياده سازي بصورت پويا و بهينه به شكل چشمگيري بهبود يافته است. نتايج تجربي بيانگر عملكرد بهتر الگوريتم پيشنهادي در ناحيه بندي تصاوير مصنوعي و پزشكي هم از نظر كيفيت پاس خها و هم از نظر زمان محاسباتي در مقايسه با چند مرز فعال متداول م يباشد.

در این مقاله روشی نوین جهت تشخیص نویسنده دستنوشته فارسی به صورت درون خط با استفاده از روش نسبتا جدید خوشه بندي ابر پارا مغناطیسی ارائه می شود. ما از این روش خوشه بندي، جهت خوشه بندي بدون ناظر و سریع دستنوشته فارسی استفاده کردیم که در مقایسه با روشهاي آزمایش شده قبلی نتایج بهتري بدست آوردیم. همچنین، علیرغم وجود ویژگیهاي متعدد، فقط به تغییر مکانهاي دستنوشته در جهتهاي تعیین شده بسنده کردیم که منجر به افزایش سرعت پیش پردازش و دسته بندي بردارهاي آموزشی شده است. همچنین از یک آستانه مبتنی بر اندازه تغییر مکان در جهتی خاص استفاده کردیم که با حذف نقاط گسستگی در یک کلمه باعث افزایش دقت دسته بندي کننده شده است. ما کارایی این روش را با استفاده از نتایج آزمایشی بر روي دستنوشته هاي فارسی نشان میدهیم.

Recent studies have shown in which extent peerto- peer file sharing traffic occurs in the global Internet. Mobile users want the same type of service, thus supporting them with peer-to-peer applications is naturally the next step. The main goal in the design of proposed architecture was to provide searching for shared files not only among vehicles during opportunistic meetings, but also among all vehicles in the Vehiclular Ad Hoc Networks. In this architecture new roles to access points are attributed. A key point in this plan is to build clusters and choose supernodes. An important criterion for any choice making no doubt is the minimization of message overhead. In this Paper the influence of incorporating vehicles traffic in the organization of access points, add to connection of distant vehicles has been studied. I have built a stable architecture providing support for a high dynamic environment beside conceiving a protocol supporting for this architecture. In addition to traditional services of indexing and searching, strategies for a cost effective car-positions management, synchronization of requests and deliveries have been developed.

Multiprocessor scheduling belongs to a special category of NP-complete computational problems. The purpose of scheduling is to scatter tasks among the processors in such a way that the precedence constraints between tasks are kept, and the total execution time is minimized. Cellular automata (CA) can be used for multiprocessor scheduling, but one of the difficulties in using CA is the exponentially increasing number of rules with increasing number of processor and neighborhood radius. Here, we propose a combined use of ant colony and evolutionary meta-heuristics to search the rule’s feasible space in order to find optimal rule base. Also we introduce a two dimensional cellular automata structure based on the important task attributes in the precedence task graph. The proposed scheduler that uses evolving cellular automata based on ant colony can find optimal response time for some of well known precedence task graph in the multiprocessor scheduling area.

Two new high-performance Full Adders, purely designed with 3-input Majority-not function, are proposed in this paper. The Majority-not function is implemented efficiently by using only capacitors and a static CMOS inverter. This kind of design improves the parameters of the Full Adder cell and leads to high performance, driving capability, a high degree of regularity and simplicity. Five state-of-the-art 1-bit Full Adder cells and the proposed Full Adders are simulated using 0.18μm CMOS technology at three supply voltages. Simulation results demonstrate significant improvement in terms of power consumption and Power-Delay Product (PDP).

Accurate pupil segmentation is the first and most important step for an iris recognition system. Current methods are based on fitting a model such as circle or ellipse to find and detect pupil, while these methods don’t have sufficient accuracy and are sensitive to the specular spot reflection. In this paper, we utilize an optimized color mapping to increase the accuracy of pupil segmentation, regardless of pupil model and its shape (circular or elliptic), while removing the effects of specular spot reflection. The optimum color mapping can be established by an iterative minimization algorithm similar to Levenberg- Marquardt (LM) method. By applying this method, a new image is provided with a clear pupil region that can be easily segmented. Also a robust preprocessing step is presented in this paper that sharpens and clears pupil region. We obtain 98% accuracy in pupil boundary detection by applying this method on UBIRIS dataset. Also, the proposed method works well on any model of eye image even where the eye is not perpendicular to the camera.

In this paper, we present a novel hybrid algorithm for binarization of badly illuminated document images. This algorithm locally enhances the document image and makes the gray levels of text and background pixels separable. Afterward a simple global binarization algorithm binarizes the enhanced image. The enhancement process is a novel method that uses a separate transformation function to map the gray level of each pixel into a new domain. For each pixel, the transformation function is determined using its neighboring pixels gray level. The proposed binarization algorithm is robust for wide variety of degraded document images. Evaluation over a set of degraded document images illustrates the effectiveness of our proposed binarization algorithm.

In real world, optimization problems are usually dynamic in which local optima of the problem change. Hence, in these optimization problems goal is not only to find global optimum but also to track its changes. In this paper, we propose a variant of cellular PSO, a new hybrid model of particle swarm optimization and cellular automata, which addresses dynamic optimization. In the proposed model, population is split among cells of cellular automata embedded in the search space. Each cell of cellular automata can contain a specified number of particles in order to keep the diversity of swarm. Moreover, we utilize the exploration capability of quantum particles in order to find position of new local optima quickly. To do so, after a change in environment is detected, some of the particles in the cell change their role from standard particles to quantum for few iterations. Experimental results on moving peaks benchmark show that the proposed algorithm outperforms mQSO, a well-known multi swarm model for dynamic optimization, in many environments.

Though numerous approaches have been proposed for face recognition. In this paper we propose a novel face recognition approach based on adaptively weighted patch Local Statistic in Multi dimensional (LMDS) when only one exemplar image per person is available. In this approach, a face image is decomposed into a set of equal-sized patches in a nonoverlapping way. In order to obtain Local Multi Dimensional Statistic Features in each patch, we calculated mean and standard deviation of all pixels along some directions. An adaptively weighting scheme is used to assign proper weights to each LMDS features to adjust the contribution of each local area of a face in terms of the quantity of identity information that a patch contains. An extensive experimental investigation is conducted using AR face databases covering face recognition under controlled/ideal conditions and different facial expressions. The system performance is compared with the performance of four benchmark approaches. The encouraging experimental results demonstrate that our approach can be used for face recognition and patch-based local statistic features provides a novel way for face.

Reversible Cellular Automata is applicable in cryptographic functions. A reversible CA can be obtained by employing reversible rules. In this paper, we propose 65280 two-dimensional reversible CA rules by compounding one-dimensional reversible rules. We produce these rules by alternately applying onedimensional CA rule f1 to the rows and applying another rule f2 to the columns of the configuration matrix of twodimensional CA. In addition to describing these rules, we consider the required cryptographic properties such as completeness, strict avalanche criteria, non-linearity, and differential-profile flatness for our proposed approach. According to the obtained results, 9463 rules are specified as appropriate rules for the purpose of applying in cryptographic functions. Only 510 rules do not satisfy any of the mentioned cryptographic properties and the other remain rules satisfy some of the properties.