Smart City Crime Detector

Inspiration

India has the highest number of road accidents in the world. This doesn’t only include road accidents but also many violent crimes such as violence, harassment, robbery, etc. Reports show that around 7.5 cases are shut down due to a lack of proper evidence. A lack of evidence can essentially put an end to a person's legal case. Not only the lack of proof delays justice but also the authorities aren’t contacted. Many times, the police department shows up after hours at the crime scene. The issue of unreported crimes is a significant concern in many societies. Addressing this issue requires efforts to create a safe environment for reporting and to educate the public about the importance of reporting crimes to ensure justice and public safety. Thus India requires a much smarter way to report crimes and provide justice to the citizens.

What it does

The solution proposes the idea to make the cameras on the streets smarter. Using technologies such as artificial intelligence, machine learning, and computer vision in the CCTV cameras on roads will make it feasible.

How we built it

The model will detect violent activity using different training data. After the detection of violent acts, the model can report to the relevant authorities the video of the crime. This will also give them hardcore evidence, and the authorities can reach the spot easily.

Challenges we ran into

One of the biggest challenges in this project is collecting and testing data for the model. Data is the foundation of this project, and obtaining relevant and accurate datasets is crucial for achieving better accuracy and predictions. It's equally important to use the correct data for both testing and training the model, as this greatly impacts the accuracy of the predictions.

Accomplishments that we're proud of

The prototype produces accurate results with high precision and recall values, making it a valuable tool for predicting outcomes.

What we learned

Through machine learning, we can classify images and sort them accordingly. Additionally, we can utilize this classification to make predictions about other images. By using a high-quality dataset and conducting thorough training and testing, a machine learning model can produce reliable results.

What's next for Smart City Crime Detector

The Smart City Crime Detector project has great potential in the market and we can enhance it by adding multiple features. For instance, we can include live location and media options. Additionally, the model can send out contact details and even the entire case to the courts. This will help to address the problem of unreported cases and make the future safer.

Built With

• ai
• cv
• machine-learning