Medicinal plant recognition manually takes a lot of time and money. Moreover, to reduce these resources, some researchers propose to implement artificial intelligence technology. This paper aims are to conduct a syste...
详细信息
ISBN:
(纸本)9781665427340
Medicinal plant recognition manually takes a lot of time and money. Moreover, to reduce these resources, some researchers propose to implement artificial intelligence technology. This paper aims are to conduct a systematic literature review of medicinal plant leaf recognition published in the last two years (2019–2020) from IEEE, Springer and Science Direct. We obtained 15 studies in the field of medicinal plant leaf recognition using artificial intelligence. The dataset used for medicinal plant leaf recognition is mostly used private dataset, however, there are public dataset named Leaf, Flavia, Swedish dataset. We also found robust method that can be used for medicinal plant leaf recognition is Multichannel Modified Local Gradient Pattern (MCMLGP) and Gray Level Co-Occurrence Matrix (GLCM) as feature extraction; and Convolutional Neural Network (CNN), Multi-Layer Perceptron trained with backpropagation algorithm (MLP-BP), Support Vector Machine (SVM), and Transfer Learning (VGG19) as classifier.
We propose a conditional generative adversarial network (GAN) incorporating humans' perceptual evaluations. A deep neural network (DNN)-based generator of a GAN can represent a real-data distribution accurately bu...
详细信息
We propose a conditional generative adversarial network (GAN) incorporating humans' perceptual evaluations. A deep neural network (DNN)-based generator of a GAN can represent a real-data distribution accurately but can never represent a human-acceptable distribution, which are ranges of data in which humans accept the naturalness regardless of whether the data are real or not. A Human-GAN was proposed to model the human-acceptable distribution. A DNN-based generator is trained using a human-based discriminator, i.e., humans' perceptual evaluations, instead of the GAN's DNN-based discriminator. However, the HumanGAN cannot represent conditional distributions. This paper proposes the HumanACGAN, a theoretical extension of the HumanGAN, to deal with conditional human-acceptable distributions. Our HumanACGAN trains a DNN-based conditional generator by regarding humans as not only a discriminator but also an auxiliary classifier. The generator is trained by deceiving the human-based discriminator that scores the unconditioned naturalness and the human-based classifier that scores the class-conditioned perceptual acceptability. The training can be executed using the backpropagation algorithm involving humans' perceptual evaluations. Our experimental results in phoneme perception demonstrate that our HumanACGAN can successfully train this conditional generator.
Formation control of multi-agent systems has been an important task in the fields of automatic control and robotics. The aim of this paper is to develop a deep learning based formation control strategy for the multi-a...
详细信息
ISBN:
(纸本)9781728100845
Formation control of multi-agent systems has been an important task in the fields of automatic control and robotics. The aim of this paper is to develop a deep learning based formation control strategy for the multi-agent systems by using the backpropagation algorithm. Specifically, the deep learning network can be treated as the feedback controller, thus the multi-agent system can use the network output as its input to achieve the formation control. The algorithm has been tested on a multirobot system to verify the effectiveness of the proposed method.
The radial basis function (RBF) network may serve as a good alternative for multilayer perceptron (MLP), since RBFN structure is much easier, and learning speed is faster. This article aims to present the architecture...
详细信息
ISBN:
(数字)9781728174433
ISBN:
(纸本)9781728174440
The radial basis function (RBF) network may serve as a good alternative for multilayer perceptron (MLP), since RBFN structure is much easier, and learning speed is faster. This article aims to present the architecture of the multilayer radial basis function network and the usage of the backpropagation algorithm as one of the approaches for finetuning the weights of a neural network based on the error rate. Using Epanechnikov kernels as activations function allows to avoid the undesirable phenomena during the tuning like vanishing and exploiding gradient. In addition, because input signal of the last layer is linearly depends of the tuning synaptic weights, this allows optimize the speed of learning process. In result error backpropagation algorithm for multilayer network was proposed, whose layers in fact are radial-basis function neural network.
Face recognition is the process of identifying the human face with the help of computerized system mostly for the purpose of security reasons. As the face is very important characteristic of human body through which o...
详细信息
ISBN:
(数字)9781728159706
ISBN:
(纸本)9781728159713
Face recognition is the process of identifying the human face with the help of computerized system mostly for the purpose of security reasons. As the face is very important characteristic of human body through which one can uniquely identify the person. So focusing the face for authentication or identification is at most important for variety of its applications. It is the problem of pattern recognition which can be easily and efficiently solved with artificial neural networks (ANN). Pattern recognition is the process through which particular patterns can be identified, recognized based on the relatedness and similarities between the patterns. Face recognition is having number of applications such as authentication at various offices, organizations or other sectors where security is of utmost important. The proposed methodology for face recognition made use of visible, thermal and fused image database. Thermal imaging detects the heat of the object or face. The artificial neural network is a very efficient technology to enhance the results. Here backpropagation and Levenberg-Marquardt algorithm were compared by using the same database. The backpropagation algorithm achieved the accuracy of 92.86% and for Levenberg-Marquardt accuracy is 83.92%.
Human action recognition is an essential topic in computer vision research that has received a wide concern. In the previous decade, deep Convolutional Neural Networks (CNN) have offered an effective execution for sev...
详细信息
ISBN:
(数字)9781728148014
ISBN:
(纸本)9781728148021
Human action recognition is an essential topic in computer vision research that has received a wide concern. In the previous decade, deep Convolutional Neural Networks (CNN) have offered an effective execution for several action recognition datasets. However, utilizing an immense amount of training data and using backpropagation algorithm in the training stage have a negative influence on its efficiency for action recognition problem. To help alleviate these limitations, we present a novel technique for human action recognition based on Principal Component Analysis Network (PCANet), which is a simple deep learning network. A subset of frames is selected from each action while for each frame a feature vector is calculated from the previously trained PCANet. All feature vectors are then fused and their dimensionality are reduced using Whitening Principal Component Analysis algorithm (WPCA). Finally, Support Vector Machines (SVM) classifier is employed for action recognition. We assess the proposed approach on the challenging dataset, KTH Human Action Dataset. Our experimental results using the leave-one-out evaluation strategy show the efficiency of the proposed method.
In this paper a new weight initialization scheme is proposed. The proposed scheme, distributes the biases / thresholds at a equal intervals in an interval ( - λ, λ); while the weights are distributed uniformly in th...
详细信息
ISBN:
(数字)9781728170169
ISBN:
(纸本)9781728170176
In this paper a new weight initialization scheme is proposed. The proposed scheme, distributes the biases / thresholds at a equal intervals in an interval ( - λ, λ); while the weights are distributed uniformly in the same interval. The value of the λ is chosen such that the expected value of the net inputs is 0 while the variance is 1. On a set of 10 tasks (5 function approximation and 5 real life benchmark regression tasks), the proposed weight initialization routine was compared to three existing weight initialization routine. From the results obtained we may infer that the proposed weight initialization is almost always better than these three existing routines and is never worse on the basis of generalization performance (result over the test data set).
backpropagation (BP) has long training time and slow convergence, i.e. in minimizing Mean Square Error (MSE). Adaptive Moment Estimation (ADAM) was employed to speed up training and to obtain rate of BP. Result shown ...
详细信息
ISBN:
(数字)9781728154237
ISBN:
(纸本)9781728154244
backpropagation (BP) has long training time and slow convergence, i.e. in minimizing Mean Square Error (MSE). Adaptive Moment Estimation (ADAM) was employed to speed up training and to obtain rate of BP. Result shown that the minimum MSE obtained was 0.002479194 and 0.002281315 in of 1000 and 2000, respectively.
To better utilize artificial intelligence (AI) in the edge domain, such that it is more attractive and fruitful, the development of low-power and -resource AI devices dedicated to online learning is very important iss...
详细信息
ISBN:
(纸本)9781728119588
To better utilize artificial intelligence (AI) in the edge domain, such that it is more attractive and fruitful, the development of low-power and -resource AI devices dedicated to online learning is very important issue to be solved. We proposed new ternarized backpropagation (TBP) algorithms [1] and verified to be favorably compatible with fixed-point 16 bit backpropagation (FixedBP). In this paper, we present our implementation method of TBP on an FPGA as an accelerator for embedded microcontrollers, and evaluate the TBP on the FPGA to achieve a 15.7 % reduction of the logic elements of the FPGA, 12.3 % reduction of the registers, 90.9 % reduction of the multipliers, and 49.8 % reduction of the SRAM usage, comparing with those for the FixedBP. We verify its capabilities for training MNIST classification task with a mini-batch size of one. In addition, we demonstrate image recognition system as application example.
There are many rice fields and rice milling factories. Rice Milling Units (RMU) are still many who have not applied the prediction method for the sale of rice so that it can affect the availability of raw materials. T...
详细信息
ISBN:
(数字)9781728154237
ISBN:
(纸本)9781728154244
There are many rice fields and rice milling factories. Rice Milling Units (RMU) are still many who have not applied the prediction method for the sale of rice so that it can affect the availability of raw materials. The purpose of this research is to forecast product sales (rice) at RMU so that it can know the number of raw materials needed so that they avoid idle time. Obtain optimal forecasting; it compared to 2 (two) forecasting methods, Linear Regression, and Artificial Neural network with a backpropagation algorithm. The results showed that the value of MSE on a linear regression method of 0.214, while at the time using Artificial Neural Network obtained an MSE value of 0.00099713. Based on the value of the MSE, the smallest MSE is forecasting by the backpropagation Neural Network method.
暂无评论