مقالات انگلیسی
Abstract
Face detection is a crucial task in computer vision and image processing with various practical applications including security, surveillance and entertainment. In recent years, approaches based on deep learning methods improved significantly with high-accuracy detection results. In this paper, we propose a new framework for anefficient face detection based on improved EfficientDet architecture and wavelet transform. Our method utilizes a combination of innovated bi-directional feature pyramid network (BiFPN) and a dual-tree complex wavelet transform called WT-BiFPN to ameliorate feature representation of faces at multiple scales. We evaluate our approach on two benchmark datasets, including WIDER FACE and our gathered dataset with specific details close to real-world images. Our proposed architecture achieves more than 5% performance improvement on previous state-of-the-art EfficientDet-based methods in terms of mean average precision (mAP). Our method provides an effective accurate face detection solution for several applications and can prevail over low resolution and occlusion in images.
Abstract
Abstract
Abstract
Low-quality face recognition (LQFR), unlike high-quality face recognition is very challenging. This is due to the standoff between the subject and the camera and small face size. The most common way to overcome the mentioned problems is using super-resolution (SR) techniques. In this paper, we show that the efficiency of a FR system degrades significantly in low-resolution images. We propose a novel FR approach using an index-based super-resolution method to improve performance and computational load of a FR system. To achieve this goal, first we apply face quality assessment (FQA) to select appropriate face images with high quality from sequence of images. Then, we introduce Blind/Reference-less Image Spatial Quality Evaluator (BRISQUE) as an index based on which we decided whether or not to use super-resolution in a LQFR system. We also conduct experiments to explore factors which affect FR performance including super-resolution methods, face image size (pixels), face quality and change of the sequence of the conventional LQFR stages. To demonstrate the efficiency of the proposed approach, we compare face identification rate on various deep CNN face recognition. Experimental results show that the proposed method increases the average identification rate to 71.20% on SCFace dataset.
Abstract
In this work, we introduce a new non-Shilnikov chaotic system with an infinite number of nonhyperbolic equilibrium points. The proposed system does not have any linear term, and it is worth noting that the new system has one equilibrium point with triple zero eigenvalues at the origin. Also, the novel system has an infinite number of equilibrium points with double zero eigenvalues that are located on the –axis. Numerical analysis of the system reveals many strong dynamics. The new system exhibits multistability and antimonotonicity. Multistability implies the coexistence of many periodic, limit cycle, and chaotic attractors under different initial values. Also, bifurcation analysis of the system shows interesting phenomena such as periodic window, period-doubling route to chaos, and inverse period-doubling bifurcations. Moreover, the complexity of the system is analyzed by computing spectral entropy. The spectral entropy distribution under different initial values is very scattered and shows that the new system has numerous multiple attractors. Finally, chaos-based encoding/decoding algorithms for secure data transmission are developed by designing a state chain diagram, which indicates the applicability of the new chaotic system.
Abstract
In this work, we present a novel three-dimensional chaotic system with only two cubic nonlinear terms. Dynamical behavior of the system reveals a period-subtracting bifurcation structure containing all mth-order (m=1,2,3,…) periods that are found in the dynamical evolution of the novel system concerning different values of parameters. The new system could be evolved into different states such as point attractor, limit cycle, strange attractor and butterfly strange attractor by changing the parameters. Also, the system is multistable, which implies another feature of a chaotic system known as the coexistence of numerous spiral attractors with one limit cycle under different initial values. Furthermore, bifurcation analysis reveals interesting phenomena such as period-doubling route to chaos, antimonotonicity, periodic solutions, and quasi-periodic motion. In the meantime, the existence of periodic solutions is confirmed via constructed Poincaré return maps. In addition, by studying the influence of system parameters on complexity, it is confirmed that the chaotic system has high spectral entropy. Numerical analysis indicates that the system has a wide variety of strong dynamics. Finally, a message coding application of the proposed system is developed based on periodic solutions, which indicates the importance of studying periodic solutions in dynamical systems.
Abstract
In this work, we present a novel three-dimensional chaotic system with only two cubic nonlinear terms. Dynamical behavior of the system reveals a period-subtracting bifurcation structure containing all mth-order ( m=1,2,3,…“>m=1,2,3,…) periods that are found in the dynamical evolution of the novel system concerning different values of parameters. The new system could be evolved into different states such as point attractor, limit cycle, strange attractor and butterfly strange attractor by changing the parameters. Also, the system is multistable, which implies another feature of a chaotic system known as the coexistence of numerous spiral attractors with one limit cycle under different initial values. Furthermore, bifurcation analysis reveals interesting phenomena such as period-doubling route to chaos, antimonotonicity, periodic solutions, and quasi-periodic motion. In the meantime, the existence of periodic solutions is confirmed via constructed Poincaré return maps. In addition, by studying the influence of system parameters on complexity, it is confirmed that the chaotic system has high spectral entropy. Numerical analysis indicates that the system has a wide variety of strong dynamics. Finally, a message coding application of the proposed system is developed based on periodic solutions, which indicates the importance of studying periodic solutions in dynamical systems.
Abstract
Pedestrian attributes recognition is an important issue in computer vision and has a special role in the field of video surveillance. The previous methods presented to solve this issue are mainly based on multi-label end-to-end deep neural networks. These methods neglect to apply attributes for defining local feature areas and they suffer from the problems of the bounding box presence. A new framework for jointly human semantic parsing and pedestrian attribute recognition to achieve effective attribute recognition is proposed. By extracting human parts via semantic parsing, both semantic and spatial information can be explored with eliminating of background. The framework also uses multi-scale features to employ rich details and contextual information through proposed attribute recognition-bidirectional feature pyramid network. For baseline network that has a significant impact on the performance, EfficientNet-B3 is selected as a baseline network from The EfficientNet family which provides an appropriate trade-off between the three factors of CNNs scaling (depth/width/resolution). Finally, the proposed framework is tested on datasets PETA, RAP and PA-100k. Experimental results show that our method has superior performance in both mean accuracy and instance-based metrics compared to state-of-the-art results.
Abstract
Background & Purpose: Cognitive performance includes a set of capacities that enable the individuals to recognize, process, and respond to information. Cognitive performance, mental structures, and cognitive functions of strategic organizations’ managers are very critical factors in the success of organizations and achieving their strategic goals. Therefore, this study aims to identify and prioritize the cognitive functions affecting the cognitive performance of strategic managers.
Methodology: The method used in this research is a qualitative-quantitative sequential mixed research method in which a combination of quantitative and qualitative methods is used sequentially. In this research, first using qualitative methods, the cognitive functions affecting managers’ cognitive performance are identified and then in a quantitative phase, using statistical analysis, a novel methods of multi-criteria decision making the identified factors are prioritized.
Findings: In the qualitative section, after reviewing the literature, 14 cognitive functions affecting the cognitive performance of managers were extracted and the conceptual model of the research is presented. In the quantitative part, using the implementation of statistical methods and a multi-criteria decision making method, cognitive functions were prioritized and the weight of each of these functions is calculated. “Planning function” and “performing several tasks simultaneously” have been selected as the most important functions among the prioritized cognitive functions, which have obtained the highest average final weight.
Conclusion: It is concluded that avoiding managerial errors due to low levels of managerial performance in any of the cognitive functions can cause irreparable problems that subconsciously reduce the efficiency of relevant organizations. In this research, by identifying and prioritizing cognitive functions affecting the cognitive performance of strategic managers, it is possible to prevent cognitive errors, reduce the accuracy factor, and increase the accuracy of decisions.
Abstract
Aim and introduction: The field of life sciences is one of the most important fields related to issues and issues affecting life, including pharmacy, therapeutic, and related technologies with direct and indirect effects on human health due to their nature. The increasing complexity of communication between sciences affecting human health in Iran. Significant scientific advances in recent years, international developments, special social conditions in Iran, and the efforts of different countries in this field, make life sciences in Iran even more important key. Also, the need for this research is clear due to the need to provide effective strategies for the development of this field in Iran. Therefore, the purpose of this article is to identify strategies for the development of life sciences and technologies in Iran. The main question of this research is focused on identifying the position of Iran in the field of life sciences, analyzing the situation in the country, and determining effective strategies to improve the current situation in the country.
Methodology: This research is categorized in terms of interpretive paradigm, type of applied goal, descriptive research method, combined or quantitative-qualitative approach, library, and field information collection. Policy-making and the presentation of strategic plans for the field of life sciences require an understanding of the current state of the country and its future needs. Therefore, in the first part of this study, the position of life sciences in Iran, the target population, and the factors affecting these sciences are determined based on field studies and the results of designed questionnaires. Then, by examining the strengths and weaknesses along with the threats and opportunities in the SWOT analysis, appropriate strategies are presented. In the following, the responsible institutions were determined and their roles for the implementation of these strategies were determined based on the principles of institutional mapping. It is also used to determine the priority of each strategy according to the experts’ knowledge and the analytic network process (ANP).
Findings: In the first stage of this study, the factors affecting the development of life sciences and technologies in Iran were examined by the target community, which included some experts and active researchers in this field. For this purpose, studies were conducted and questions were designed based on the SERVQUAL model. The focus of the main questions was determined based on the relevant specialties and current knowledge due to the statistical sample with different disciplines and specialties. After designing the main questions in the questionnaires, their validity was assessed with the content validity ratio index and their reliability was assessed by Cronbach’s alpha method. After ensuring the validity and reliability of the questionnaires, they were analyzed based on the SWOT method. According to this analysis, the most important strength, weaknesses, opportunities, and threats are coordination of the country’s scientific services and products with international needs; becoming this field as a bridge to leave the country due to the knowledge frontier; Iran’s richness in raw materials, organic and plant structures; focus on producing articles without considering their application, respectively. Other identified items were ranked according to their frequency and importance. Then, by dividing the strengths and weaknesses related to Iran in the field of life sciences and confronting them with identified opportunities and threats, solutions were extracted. These solutions are designed based on different aspects including managerial, political, technical, infrastructural, social, and communication perspectives. Also, the institutions related to the field of life sciences were identified and the proposed solutions were assigned to them using institutional maps according to the results obtained from the completed questionnaires. Based on the results of the questionnaires, the weight and importance of each solution were determined and their priority was determined based on the analytic network process (ANP).
Discussion and Conclusion: According to the results of this study, the most effective strategy to increase the relationship between science and industry is to hold international conferences and scientific meetings in the field of life sciences and the convergence of related sciences in this field to monitor developments in different countries. This strategy can attract domestic and foreign investment and flourish many research projects. The next strategy is to determine the regulating laws of capital between different fields of life sciences and technologies in Iran according to national and international needs and trends, which will lead to the balanced development of the country in this field. Also, the development of centers for organizing materials and logistics related to the field of life sciences in Iran is the next effective strategy that can stabilize the research processes of many companies active in this field. On the other hand, investment policies in the country should be organized in accordance with the needs of the country so that in addition to the country’s self-sufficiency in this area, foreign markets should be targeted. The results of this research should be reviewed regularly to identify the most appropriate strategies in any situation and time because the dynamics of current scientific developments is one of the limitations of the present study.
Abstract
Face recognition domain has been well advanced and has achieved high accuracies in identification of individuals in recent years. But in practice, distinguishing similar faces such as an identical twin still is a great challenge for face recognition systems. It happens due to very small differences in the facial features of them. Therefore, extracting common face features is not proper for differentiating identical twins. A solution to this problem is to find the most distinctive regions in the face of identical twins. In this paper, two procedures used to find these specific regions: 1) Machine Processing: A Modified SIFT (M-SIFT) algorithm has been implemented on Identical twins’ face images. Each face image has been segmented into five regions contain eyes, eyebrows, nose, mouth, and face curve. The location and number of mismatched keypoints represented the most distinctive face region in the face of identical twins. 2) Crowdsourcing: We have recognized differences between identical twins faces from human criteria viewpoint by enlisting crowd intelligence. Several questionnaires were designed and completed by 120 participants. The dataset of this study collected by ourselves and include 650 images for 115 pairs of identical twins and 120 non-twin individuals. The results of Machine Processing and Crowdsourcing methods showed that the face curve is the most discriminant region among every five regions in most of identical twins. Several features proposed and extracted based on the keypoints of the M-SIFT algorithm and face landmarks. The experimental results demonstrated the lowest equal error rate of identical twins recognition as 7.8, 8.1 and 10.1% for using the whole images, only frontal images and only images with PAN motions, respectively.
Abstract
Human detection in images is a crucial task due to its usage in different areas including person detection and identification, abnormal surveillance and crowd counting. Low-resolution of image sequences taken by stationary outdoor surveillance cameras is very challenging. Detecting human with deep learning techniques, is more powerful than traditional methods due to its ability to learn high-level deeper features, high detection accuracy and speed. Therefore, this paper proposes a method for human detection in low-resolution images based on YOLOv3. This method will prepare a dataset of low-resolution images collected by outdoor surveillance cameras and annotate them manually. Next, we retrain YOLOv3 to make an improved model for low-resolution images. The model achieves F1-score of 0.804 human detecting for low-resolution test images.
Abstract
Introduction: Wearable electronic devices, which are based on Internet of Things (IoT) and big data computing, are able to continuously collect and process the physiological and environmental data and exchange them with other tools, users, and internet networks. Therefore, despite their potential benefits in health monitoring, they can pose serious risks, especially in breach of privacy. Hence, the main question in this study was to identify the most important strengths, weaknesses, opportunities, and threats related to wearable electronic technologies.
Method: In this study, StArt 3.4 software was used for systematic review. Studies until November 30, 2019 were searched for keywords in “Scopus”, “IEEE”, “PubMed”, “Springer”, “Magiran”, “SID”, and “Sivilica” databases and Google search engine.
Results: After deleting duplicate and unrelated documents, 80 documents were selected for final review and were analyzed using descriptive statistics. Accordingly, the main identified strength, weakness, opportunity, and threat were “improving lifestyle and human capabilities”, “low data reliability and user interface”, “applications in health and medicine”, and “information abuse and privacy breach” with 97.5%, 92.5%, 94%, and 99% frequency, respectively.
Conclusion: The results of this study showed that improving human capabilities and application in medicine and health care are the main driving forces for the development of wearable electronic technologies. Therefore, in order to take advantage of the opportunities and overcome the potential threats of this technology, planning for the development and application of indigenous knowledge, as well as the development of the required standards and rules, must be put on the agenda immediately.
Abstract
Cognitive warfare, in which, using human’s brain cognitive biases, enemy is intending to disturb people’s cognitive abilities, is one type of psychological warfare. In this research, gathering questionnaire, we found that the most important cognitive error used in cognitive warfare in exchange market is Herding; in this line, we try to define an index to measure cognitive warfare intensity in and its impact on exchange market. Using the results of estimations, we predicted how exchange rate will differ.
Our results shows that if, taking advantage of real exchange rate index, we control for the inflation rate in Iran and its trade partner and major countries in world trade and also control for monetary policy shocks, herding index, which is used as a proxy for cognitive warfare, could account for almost 50 percent of changes in USD exchange rate. In 45 percent of months the prediction from exchange rate babble by cognitive warfare intensity in previous month have less than one standard deviation error. Moreover, the results of prediction of exchange rate have in more than a half on months have less than 5 percent error. These results are fruitful for preparing against psychological attacks in exchange market, and anti-psychological warfare operations.
Abstract
The field of neurotechnology has a bright future ahead of it due to its direct impact on human health and the various technological applications. Consequently, this field is of great importance to many developed countries. As far as we know, since most of the neurotechnologies are emerging technologies, there is no proper strategic analysis for this field in the country. The purpose of this study is to determine the technology readiness level (TRL) for the branches of neurotechnology and identify and prioritize the appropriate strategies by evaluating the current state of the country in the neurotechnology field. To achieve this goal, questionnaires are distributed among 40 experts in neuroscience and neural technologies. In the strategy prioritization and data analysis step, the SWOT matrix-based method is used; then, the weights of internal and external factors are determined. After performing a pairwise comparison for selecting the top 5 strategies, a quantitative strategic planning matrix (QSPM) is used in order to rank the selected ones.
The results indicate that Iran has the highest technology readiness level (level 6) in the brain-to-computer interfaces and the lowest technology readiness level (level 1) belongs to brain-to-brain interfaces. Finally, 5 top strategies which are weakness- opportunity (WO) strategies are determined.
Abstract
Abstract
A physiologically realistic three layer neuron-astrocyte network model is used to evaluate the biological mechanism in pattern separation. The innovative feature of the model is the use of a combination of three elements: neuron, interneuron and astrocyte. In the input layer, a pyramidal neuron receives input patterns from stimulus current, while in the middle layer there are two pyramidal neurons coupled with two inhibitory interneurons and an astrocyte. Finally, in the third layer, a pyramidal neuron produces the output of the model by integrating the output of two neurons from the middle layer resulting from inhibitory and excitatory connections among neurons, interneurons and the astrocyte. Results of computer simulations show that the neuron-astrocyte network within the hippocampal dentate gyrus can generate diverse, complex and different output patterns to given inputs. It is concluded that astrocytes within the dentate gyrus play an important role in the pattern separation process.
Abstract
Methods: The statistical population in this study included all managers and staff of the National Defense University in the first half of 2019, of which 189 people were evaluated. The Military Managers Strategic Thinking Test was designed by Weyhrauch (2017) and under the auspices of the US Army Institute for Military Studies in Behavioral and Cognitive Sciences. The test consists of 12 scenarios in which the respondent chooses one answer in each scenario. The Goldman Strategic Thinking Questionnaire (2007) was also used to assess criterion validity.
Results: Analysis of research data showed that the structure based on three subscales using the confirmatory factor analysis to explain and fit a good situation. Also, the reliability coefficients of internal consistency and test-retest were optimal. Accordingly, the researcher’s hypothesis that the structure is based on three components of intellectual inclusiveness, intellectual humility, and intellectual flexibility is confirmed.
Conclusion: It seems that research evidence supports the consideration of all three components as cognitive dimensions for strategic military thinking, and this test can be used in military assessments.
Abstract
Abstract
The nervous system is known as a crucial part of the body and derangement in this system can cause potentially lethal consequences or serious side effects. Unfortunately, the nervous system is unable to rehabilitate damaged regions following seriously debilitating disorders such as stroke, spinal cord injury and brain trauma which, in turn, lead to the reduction of quality of life for the patient. Major challenges in restoring the damaged nervous system are low regenerative capacity and the complexity of physiology system. Synthetic polymeric biomaterials with outstanding properties such as excellent biocompatibility and non-immunogenicity find a wide range of applications in biomedical fields especially neural implants and nerve tissue engineering scaffolds. Despite these advancements, tailoring polymeric biomaterials for design of a desired scaffold is fundamental issue that needs tremendous attention to promote the therapeutic benefits and minimize adverse effects. This review aims to (i) describe the nervous system and related injuries. Then, (ii) nerve tissue engineering strategies are discussed and (iii) physiochemical properties of synthetic polymeric biomaterials systematically highlighted. Moreover, tailoring synthetic polymeric biomaterials for nerve tissue engineering is reviewed.
Abstract
Measurement of semantic similarity plays an important role in many areas of natural language processing. Several approaches have been proposed to determine the similarity of sentences in different languages but many of them are not extendable in all languages. According to the complicated Arabic language structure and lack of necessary resources and tools, the Arabic semantic similarity measurement is challenging. In this paper, we proposed a supervised method for Arabic semantic question similarity measurement. Forty-one features (lexical, syntactic and semantic) are extracted from two question phrases, then the best distinctive features are selected by using SelectKBest algorithm. Finally, for sentences classification and determining the similarity score, SVM used. The system participated in task8 of NSURL 2019 .The results of using this method on the data set of NSURL 2019 have a F-measure of 82.58 percent, which have improved the basic method.
Abstract