Recognition of student emotion based on matrix-1 median fisher's face and backpropagation algorithm

Wibawanto, Slamet; Kirana, Kartika Candra

doi:10.1109/iceltics.2017.8253255

Cited by 13 publications

(11 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The original Viola-Jones system works faster because it uses simple summation based features, while the features we use work twice in PCA and LDA calculations. however, our system is more accurate because the fisher face is able to produce the best vector dimension to represent the face and emotions as described by [9] [10] and [11]. The example of comparison results in our tests is shown in Figure 2.…”

Section: Resultsmentioning

confidence: 99%

“…LDA project the optimal matrix under Fisher criterion, but the dimension of the input space is greater than the number of training images, thus it cannot be applied directly. [9] [10] [11] Projection PCA of the matrix is computed by Equation 1 and projection LDA of the matrix is computed by Equation 2.…”

Section: Fisher Facementioning

confidence: 99%

“…However, 2D fisher face computes slower than PCA. [11] implements artificial neural networks and fisher face features to recognize emotions in the learning context. They reach an accuracy of 81%.…”

Section: Introductionmentioning

confidence: 99%

“…The above research [9][10] [11] have been shown that the fisher face feature is a robust feature in the facial emotion recognition. However, they consume a longer time than the primitive features.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Emotion Recognition using Fisher Face-based Viola-Jones Algorithm

Kirana¹,

Wibawanto²,

Herwanto³

2018

EECSI

Self Cite

View full text Add to dashboard Cite

In the form of the image integral, this primitive feature accelerates the performance of the Viola-Jones algorithm. However, the robust feature is necessary to optimize the results of emotion recognition. Previous research [11] has shown that fisher face optimized projection matrix in the low dimensional features. This feature reduction approach is expected to balance time-consuming and accuracy. Thus we proposed emotion recognition using fisher face-based Viola-Jones Algorithm. In this study, PCA and LDA are extracted to get the fisher face value. Then fisher face is filtered using Cascading AdaBoost algorithm to obtain face area. In the facial area, the Cascading AdaBoost algorithm re-employed to recognize emotions. We compared the performance of the original viola jones and fisher face-based viola jones using 50 images on the State University of Malang dataset by measuring the accuracy and time-consuming in the fps. The accuracy and time-consuming of the Viola-Jones algorithm reach 0.78 and 15 fps, whereas our proposed methods reach 0.82 and 1 fps. It can conclude that the fisher face-based viola-jones algorithm recognizes facial emotion as more accurate than the viola-jones algorithm.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Fisher Facementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Emotion Recognition using Fisher Face-based Viola-Jones Algorithm

Kirana¹,

Wibawanto²,

Herwanto³

2018

EECSI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Based on the explanation above, artificial neural networks find the best weight of each facial feature (input) iteratively [11]. However, the original neural network cannot remove non-face areas before the process is complete [6]. Thus we buried the ground between neural network functions to speed up the filter process.…”

Section: Introductionmentioning

confidence: 99%

The Improved Artificial Neural Network Based on Cosine Similarity in Facial Emotion Recognition

Kirana¹,

Wibawanto²,

Hidayah³

et al. 2019

EECSI

Self Cite

View full text Add to dashboard Cite

In this study, we present the improved artificial neural network based on cosine similarity in facial emotion recognition. We apply a shifting window that employs a neural network for two concurrent processes consisting of face detection and emotional recognition. To prevent the slow and futile computations, non-face areas need to be filtered from neurons on each network layer, thus we propose the improved artificial neural network based on cosine similarity. Cosine similarity is employed to bypass the process of non-face areas in the neural network. The accuracy of the proposed method reaches 0.84, while the accuracy of the original neural network method reaches 0.74. It can be concluded that our methods work accurately. It can be concluded that our method works accurately. The proposed method is superior to the state-of-theart algorithms

show abstract