دانلود رایگان مقاله انگلیسی تشخیص عمل براساس شبکه بیزی دینامیکی سلسله مراتبی - اشپرینگر 2018

عنوان فارسی
تشخیص عمل براساس شبکه بیزی دینامیکی سلسله مراتبی
عنوان انگلیسی
Action recognition based on hierarchical dynamic Bayesian network
صفحات مقاله فارسی
0
صفحات مقاله انگلیسی
14
سال انتشار
2018
نشریه
اشپرینگر - Springer
فرمت مقاله انگلیسی
PDF
نوع مقاله
ISI
پایگاه
اسکوپوس
کد محصول
E8930
رشته های مرتبط با این مقاله
مهندسی کامپیوتر، فناوری اطلاعات
گرایش های مرتبط با این مقاله
هوش مصنوعی، شبکه های کامپیوتری
مجله
ابزارهای چندرسانه ای و برنامه های کاربردی - Multimedia Tools and Applications
دانشگاه
Department of Electronic Information Engineering - Xi’an Technological University - People’s Republic of China
کلمات کلیدی
تشخیص عمل، HDBN، شبکه های عصبی عمیق، HASD، مدل گراف
doi یا شناسه دیجیتال
https://doi.org/10.1007/s11042-017-4614-0
۰.۰ (بدون امتیاز)
امتیاز دهید
چکیده

Abstract


In this paper, a novel action recognition method is proposed based on hierarchical dynamic Bayesian network (HDBN). The algorithm is divided into system learning stage and action recognition stage. In the stage of system learning, the video features are extracted using deep neural networks firstly, and using hierarchical clustering and assisting manually, a hierarchical action semantic dictionary (HASD) is built. The next, we construct the HDBN graph model to present video sequence. In the stage of recognition, we first get the representative frames of unknown video using deep neural networks. The features are inputted into the HDBN, and the HDBN inference is used to get recognition results. The testing results show the proposed method is promising.

نتیجه گیری

5 Conclusions


In this paper, a novel the HDBN-based action recognition method is proposed. Our contribution can be described as:


(1) We propose a novel graph-based action recognition model. The model combines the hierarchical action semantic dictionary and Bayesian graph model inference together, and uses recursion-based method to recognize action video data.


(2) Based on some theories, such as Bayesian rules, graph model, the probability-based recursion calculation structure is presented to obtain the higher accuracy of action recognition. Experimental results show that the proposed model has better performance than some existing algorithms.


بدون دیدگاه