Facial expression recognition based on a mlp neural network using constructive training algorithm

Boughrara, Hayet; Chtourou, Mohamed; Amar, Chokri Ben; Chen, Liming

doi:10.1007/s11042-014-2322-6

Cited by 89 publications

(47 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many previous approaches for gesture recognition, such as the literature [13], used Kinect's deep information to locate human hands, and detect the location of five fingers based on contour and convex concave points. The static gesture can be identified by calculating the angle characteristics of the three points.…”

Section: Related Workmentioning

confidence: 99%

“…Subsampling is to further reduce the amount of data and reduce the spatial resolution of the input image, which is to sample the input feature map [12][13]. By extracting the features of the feature map, the down sampling can be implemented simply, so as to ensure the robustness and distortion of the model's displacement and scaling.…”

Section: Fig 5 Lenet-5 Network Structurementioning

confidence: 99%

“…. First, define M Markov chain, the jointly distribution under steady state is the follow: In the algorithm, we use the (13) and odd even exchange rules to complete the B and A exchange, so that we can often get better results.…”

Section:  mentioning

confidence: 99%

See 2 more Smart Citations

Jointly network: a network based on CNN and RBM for gesture recognition

Cheng

Sun

et al. 2018

Neural Comput & Applic

View full text Add to dashboard Cite

Hand belongs to non-rigid objects and is rich in variety, making gesture recognition more difficult. The essence of dynamic gesture recognition is the classification and recognition of single-frame still images. Therefore, this paper mainly focuses on static gesture recognition. At present, there are some problems in gesture recognition, such as accuracy, real-time or poor robustness. To solve the above problems, in this paper, the Kinect sensor is used to obtain the color and depth gesture samples, and the gesture samples are processed. On this basis, a jointly network of CNN and RBM is proposed for gesture recognition. It mainly uses superposed network of multiple RBMs to carry out unsupervised feature extraction and combined with supervised feature extraction of CNN. Finally, these two features are combined to classify them. The simulation results show that the proposed jointly network has a better performance in identifying simple background gesture samples, and the recognition capability of gesture samples in complex background needs to be improved.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Fig 5 Lenet-5 Network Structurementioning

confidence: 99%

See 1 more Smart Citation

Jointly network: a network based on CNN and RBM for gesture recognition

Cheng

Sun

et al. 2018

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Boughrara [3] applied neural network for facial expression recognition. Furthermore, some researchers have improved ANN by combining it with other useful approaches.…”

Section: Related Workmentioning

confidence: 99%

Extraction of target region in lung immunohistochemical image based on artificial neural network

Fan

Cao

2016

Multimed Tools Appl

View full text Add to dashboard Cite

Immunohistochemistry is widely used in clinical pathology analysis and diagnosis, the target regions segmentation is the key procedure and always provides support for many qualitative and quantitative analyses on digitized immunohistochemical image. In lung tissue immunohistochemistry applications, the target region needs to be extracted out of the whole image firstly. Most existing methods based on color cannot fulfill the extraction of antibody region. Thus, there is a need of effective extraction method. Methods: According to the features of target region in images to be processed, this paper presents a solution framework based on artificial neural network (ANN). Results: Six effective features of the candidate regions are analyzed and extracted as the inputs of the ANN; three-layers back propagation neural network with six inputs and one output is constructed, and ANN's parameters are trained by the learning image set. By the trained ANN, target region core are obtained and then expanded to the whole target region through conditional expansion. Conclusion: Through testing the framework by testing image set and comparing with the main existing methods, it can be concluded that the proposed framework can remove non-target regions and extract the target regions well, while the contrast methods cannot remove all the non-target regions. Significance: The method presented in this paper has practical and potential significance to realize automated and quantitative tissue immunohistochemical image analysis.

show abstract

“…MLP is an improvisation of the standard linear Perceptron and can distinguish data that are not linearly separable [ 31]. Hayet Boughrara ·Mohamed Chtourou Chokri Ben Amar and Liming Chen [32] in their study of Facial expression recognition based on a MLP neural network using constructive training algorithm works with MLP of Three layer. The number of input neurons is equal to the size of related feature vector.…”

Section: Multilayer Perceptron (Mlp)mentioning

confidence: 99%