Security Enhancement in ATMs through Helmet Wear Detection

Implementation Details:
1. We are using Object detection method here.
2. We used yolo data set for detecting helmet object
3. The aim of project is giving security alert to bank manager if a person wears a helmet inside ATM machine room
4. The person may carry a helmet also.
5. But we defined our problem of identifying helmet if wear by a person
6. For this we considered finding skin inside the helmet
7. We define a threshold value for skin (if he hold in hands we will not considered as the score value for skin is less) and we detect it
8 SMS alert sent to manager in identification time

Python demo

Luggage detection and approval in Airports using CBIR (Content based Image retrieval)

1. We have collected luggage dataset from google
2. We are Extracting the features and stored in file called “dataset.csv”
3. Input image is taken
4. Find the object using Coco model (object detection), we identified Luggage objects
5. Cropped the image and stored separately
6. Object is compared with the data set features and using CBIR, we arrive score level
7. If the objects score are <=10, we approved the luggage
8. Similarly done for video
9. Video is taken frame by frame and for every frame it is identified.

Python Demo

Aspect based sentiment analysis using deep learning

Python Demo

Python demo

Named Entity Recognition for Hindi-English Code-Mixed Social Media Text

Named Entity Recognition (NER) is a major task in the field of Natural Language Processing (NLP), and also is a subtask of Information Extraction. The challenge of NER for tweets lies in the insufficient information available in a tweet. There has been a significant amount of work done related to entity extraction, but only for resource-rich languages and domains such as the newswire. Entity extraction is, in general, a challenging task for such an informal text, and code-mixed text further complicates the process with it’s unstructured and incomplete information. We propose experiments with different machine learning classification algorithms with word, character and lexical features. The algorithms we experimented with are Decision tree, Long Short-Term Memory (LSTM), and Conditional Random Field (CRF). In this paper, we present a corpus for NER in Hindi-English Code- Mixed along with extensive experiments on our machine learning models which achieved the best f1-score of 0.95 with both CRF and LSTM.

Python Demo

Python demo

Handwritten character recognition is much evolving technique. In this work, dataset is trained with SVM model. Capital Letters and Small letters of Alphabets are recognized suing opencv.

Python Demo