Facial Emotion Detection Using Haar-Cascade Classifier and Convolutional Neural Networks

Bu çalışmanın amacı koronavirüsün yayılım hızını düşürmede önemli bir etkisi olan maske takma, sosyal mesafe ve kalabılık analizinin yapılmasıdır. Bu analiz için çalışmada derin öğrenme tabanlı yöntemler olan Evrişimli Sinir Ağları (ESA) ve YOLO mimari tasarımları kullanılmıştır. Maske tespitinin yapılması için yeni bir veri seti oluşturulmuştur. Oluşturulan veri seti ‘maskeli’ ve ‘maskesiz’ sınıflandırma işleminin yapılması için AlexNet, DenseNet, MobileNet, ResNet, ShuffleNet, SqueezeNet, VGG, Xception ve ZFNet gibi ESA mimari tasarımları ile kullanılmıştır. En iyi sınıflandırma performansını %96.86 doğruluk oranı ve %91.81 F1-skoru değeri ile DenseNet-121 mimari tasarımı elde etmiştir. Sosyal mesafe ve kalabalık analizi için çalışmada YOLOv3 algoritması ve COCO veri seti kullanılmıştır. Daha sonra maske sınıflandırması görevinde elde edilen ağırlık dosyası, Haar Cascade yüz sınıflandırıcı algoritması ile birlikte kullanılarak, sosyal mesafe ve kalabalık analizini sağlayan algoritmaya dahil edilmiştir. Sonuç olarak ise hem maske tespitini sağlayan hem de sosyal mesafe ve kişi sayısını hesaplayan tümleşik bir yazılım oluşturulmuştur.

Download Full-text

Baby Cry Detection in Domestic Environment using Convolutional Neural Networks

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.g5260.059720 ◽

2020 ◽

Vol 9 (7) ◽

pp. 793-795

Keyword(s):

Neural Network ◽

Neural Networks ◽

Convolutional Neural Networks ◽

Network Architecture ◽

Web Application ◽

Well Being ◽

Emotion Detection ◽

Neural Network Architecture ◽

Domestic Environment ◽

The Neural Network

In this paper we will identify a cry signals of infants and the explanation behind the screams below 0-6 months of segment age. Detection of baby cry signals is essential for the pre-processing of various applications involving crial analysis for baby caregivers, such as emotion detection. Since cry signals hold baby well-being information and can be understood to an extent by experienced parents and experts. We train and validate the neural network architecture for baby cry detection and also test the fastAI with the neural network. Trained neural networks will provide a model and this model can predict the reason behind the cry sound. Only the cry sounds are recognized, and alert the user automatically. Created a web application by responding and detecting different emotions including hunger, tired, discomfort, bellypain.

Download Full-text

Facial Emotion Recognition for Autism Children

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.f3772.059720 ◽

2020 ◽

Vol 9 (7) ◽

pp. 1274-1278

Keyword(s):

Facial Expressions ◽

Autism Spectrum ◽

Facial Emotion ◽

Expression Recognition ◽

Emotion Detection ◽

Age And Gender ◽

Social Media Platforms ◽

Haar Cascade ◽

And Gender ◽

Cascade Method

Nowadays Autism children find it difficult to interact socially with people emotions and make themselves isolated. This paper proposes Emotion detection for Autism spectrum disorder children (ASD). It is self-possessed of python libraries Open CV, Haar-cascade method and Age and gender prediction. Conversely, most existing methods rely on the detection of facial expressions of people in social media platforms such as snapchat use facial recognition technology and also detecting facial emotions from their Facial expressions in image. And for a better involvement of the children’s social behaviour, here a face is captured in real time and age, gender and emotions are predicted by Facial expression recognition (FER). This proposed system helps to improve the Autism children behaviour as they often observe the facial expressions of humans and try to imitate their emotions which make a huge difference in their behaviour.

Download Full-text