期刊名称：International Journal of Early Childhood Special Education
电子版ISSN：1308-5581
出版年度：2022
卷号：14
期号：5
页码：1591-1603
DOI：10.9756/INTJECSE/V14I5.161
语种：English
出版社：International Journal of Early Childhood Special Education
摘要：According to the NRCB, a report submitted in the year 2020 a total of 1,08,564 missing children cases are filed. The percentage of tracing the missing children is very low. India is the second most populated nation with most of its number in youth and the number of missing children cases is increasing. This paper boons an innovative use of deep learning models for finding informed missing children from the pictures of the system database, using face recognition. The community is allowed to upload snapshots of a doubtful child into a community portal with found location and contact details. The snapshot will be accordingly compared with the system database of the registered missing children and the photo with the matched facial features will be selected from the system database and displayed. The Convolutional Neural Network (CNN), is specially designed and developed to process pixel data which helps in face recognition and processes the data for face recognition. The CNN is suitably trained to find the missing kid from the database using face recognition. CNN uses multilayer perceptron for reduced processing requirements. The neuron layers are organized in such a way as to cover a detailed visual field excluding the image processing problem of the existed traditional neural networks. This algorithm uses a convolution network as a high-end feature extractor and the child recognition is taken care of by the trained K- NN classifier. The K-Nearest Neighbour is a lazy learner algorithm that also has no Parameters unlike the other algorithm on the data. The K-NN set does not learn from the training It stores the data in the event of classification. Later, It performs the acton the algorithm uses CNN's convolutional layer. It keeps the dataset and classifies the data into a category when new data is received that is much similar to the received data and the network for feature extraction. child matching is talien care by the K-NN Classifier.It stores the dataset and classifies the data into a category when new data is received that is much similar to the received data.
