چکیده مقاله
Without using a complex Age/Gender estimate system there is a probability that an autopilot taxi ask whereI have to go sir from a female traveler CNN has a desired performance in many image processing jobs aswell as Age and Gender estimation Network, its accuracy has a limit which researches try to improve byoptimizing the network parameters, neuron type, lose functions, averaging result and so on In our case wefound that the Standard CNN Definition itself is one of the limitation cap, The CNN Layer s Architectureconsists of a number of convolutional and subsampling layers optionally followed by fully connected layers,other CNN definition that are being used are same in architecture too, but replacing all the subsamplinglayers with another overall convolution layer did the job with the cost of double training time We reached%2 5 & 4% more accuracy in Age & Gender estimate CNN
کلیدواژهها
نویسندگان
Saeed H.Zadeh
Department of Artificial Intelligence, Faculty of Engineering, Kazeroon Branch, Islamic Azad University, Kazeroon, Iran
Ahad Salimi
Faculty of Islamic Azad University, Zarindasht branch
شیوه ارجاع
H.Zadeh, Saeed and Salimi, Ahad,1397,Convolutional Neural Network New Architecture lead to better Performance on Age/Gender estimate network,Second National Conference on Computer, Information Technology and AI Applications,Ahvaz
ارائهشده در
مجموعه مقالات دومین کنفرانس ملی کامپیوتر، فناوری اطلاعات و کاربردهای هوش مصنوعی1 اسفند 1397 · اهواز