Missing child identification system

mi

M             Missing Child Identification System    

           India is the second populous country in the world and children represent a significant percentage of total population. The future of any country depends upon the right upbringing of its children. But unfortunately, a large number of children go missing every year in India due to various reasons including abduction or kidnapping, run-away children, trafficked children and lost children. A deeply disturbing fact about India’s missing children is that while on an average 174 children go missing every day, half of them remain untraced. Children who go missing may be exploited and abused for various purposes. As per the National Crime Records Bureau (NCRB) report which was cited by the Ministry of Home Affairs (MHA) in the Parliament more than one lakh children (1,11,569 in actual numbers) were reported to have gone missing till 2016, and 55,625 of them remained untraced till the end of the year. Many NGOs claim that estimates of missing children are much higher than reported. The child missing from one region may be found in another region or another state, for various reasons. So even if a child is found, it is difficult to identify him/her from the reported missing cases.

 

          2.       MOTIVATION

 

The motivation behind developing a missing child identification system using deep learning and multiclass SVM lies in the urgent need to address the pressing issue of child abduction and disappearance. Every year, countless children go missing globally, causing immense distress to families and communities. Traditional methods of identification, such as fingerprints and DNA, often prove inadequate in cases where these records are not available or have been tampered with. By leveraging the power of deep learning, which excels at extracting intricate patterns from large datasets, and multiclass SVM, which efficiently categorizes data into multiple classes, this system aims to revolutionize the process of identifying missing children.

 

Deep learning algorithms, particularly convolutional neural networks (CNNs), possess the capability to learn complex features from images, making them invaluable in analyzing photographs and surveillance footage for potential matches with missing children. Through extensive training on diverse datasets, these models can discern subtle facial characteristics and variations, enabling accurate recognition even in cases of aging or disguise. Furthermore, multiclass SVM provides a robust framework for classifying these identified individuals into various categories, aiding law enforcement agencies in prioritizing and strategizing search efforts. Additionally, because of its scalability and adaptability, it may be implemented widely, encouraging cooperation between communities and authorities around the globe to protect the most vulnerable members of society. Essentially, the motivation behind this creative approach is a collective commitment to guarantee each child's safety and wellbeing, and overcoming challenges to bring them back.

[1]         FACE  RECOGNITION  USING  HISTOGRAMS  OF ORIENTED GRADIENTS

                       

Scale Invariant Feature Transform (SIFT) has shown to be a powerful technique for general object recognition/detection. In this paper, we propose two new approaches: Volume-SIFT (VSIFT) and Partial-Descriptor-SIFT (PDSIFT) for face recognition based on the original SIFT algorithm. We compare holistic approaches: Fisher face (FLDA), the null space approach (NLDA) and Eigenfeature Regularization and Extraction (ERE) with feature- based approaches: SIFT and PDSIFT. Experiments on the ORL and AR databases show that the performance of PDSIFT is significantly better than the original SIFT approach. Moreover, PDSIFT can achieve comparable performance as the most successful holistic approach ERE and significantly outperforms FLDA and NLDA.    

    PROBLEM DEFINITION

 To create a system for identifying missing children using multiclass SVM and deep learning. In spite of differences in appearance, this involves correctly identifying absent children from visual data. Also, facial appearance of child can vary due to changes in pose, orientation, illumination, occlusions, noise in background etc. The image taken by public may not be of good quality, as some of them may be captured from a distance without the knowledge of the child. A deep learning architecture considering all these constraints is designed here.

                Proposed System

                    ·           This project proposes a method that uses a deep learning methodology for                   identifying  the reported missing child from the photos of multitude of children available, with t                  face recognition.

·       The public can upload photographs of suspicious child into a common portal with landmarks and remarks. The photo will be automatically compared with the registered photos of the missing child from the repository.

·       Classification of the input child image is performed and photo with best match will be selected from the database of missing children.

·       For this, a deep learning model is trained to correctly identify the missing child from the missing child image database provided, using the facial image uploaded by the public.

        HARDWARE REQUIREMENTS

·       Computer Desktop or Laptop

·       System will be using Processor: intel i5 or Higher

·       Main Memory: 4 GB RAM

·       Hard Disk      : 1 TB

·       Display           : 14" Monitor (For more comfort)

·       Mouse             : Optical Mouse

 

      SOFTWARE REQUIREMENTS

·       Operating  system          : Windows 10

·       Programming Language : Python 3.8

·       Front End                        : HTML, CSS, JavaScript.

·       Data Base                       : MySQL

·       IDE                                 : VSCODE

      Working of Deep Learning:

  ARCHITECHTURE DIAGRAM

              

CONCLUSION

               In this paper, A missing child identification system is proposed, which combines the powerful CNN based deep learning approach for feature extraction and support vector machine classifier for classification of different child categories. This system is evaluated with the deep learning model which is trained with feature representations of children faces. By discarding the softmax of the VGG-Face model and extracting CNN image features to train a multi class SVM, it was possible to achieve superior performance. Performance of the proposed system is tested using the photographs of children with different lighting conditions, noises and also images at different ages of children. The classification achieved a higher accuracy of 99.41% which shows that the proposed methodology of face recognition could be used for reliable missing children identification.