View All Documents & Correspondence
Snake Detector And Alerting Gadget For Rural India Using Yolo
Abstract: This invention is related to the approximate 4:3 counters based on K-Map reduction techniques. Two various approximate 4:3 counter namely, Design 1 approximate 4:3 counter & Design 2 approximate 4:3 counter are proposed. The original exact 4:3 counter has been taken as the reference, from which both of the designs are built from the scratch by using K-Map reduction and Boolean expression reduction techniques. Design 1 is derived from the exact 4:3 counter where the output 01 is only approximated to 01', however, the outputs 02 & 00 are left untouched. Design 2 is derived from the Design 1, and from the Design 1, the approximated output 01! has been taken, furthermore, the outputs 02 and 00 are also approximated to 02' and 00' with an emphasize on EX-OR gate elimination for the transistor count reduction and Error Distance reduction. Design 1 attained 75% pass rate, 30% reduction in area with the. Error Distance ranging from -2 to +2. Design 2 attained 56.25% of pass rate, 50% reduction in area with the Error Distance ranging from -1 to +1. Also it is noteworthy that the critical path delay of the Design 2 is lesser than that of the Design 1.
Notices, Deadlines & Correspondence
Patent Information
Applicants
1. VIT-AP UNIVERSITY
Inventors
1. Dr S KARTHIKEYAN
2. K CHARVI
3. NIVAS MANDUVA
4. ADITYA MITRA
5. ANIL KUMAR
Specification
Our portable snake detector aims to have an immediate response in alerting the surrounding, to have this a faster
algorithm YOLO is exploited to achieve the desirable results. Employing lot devices for a completely independent
functional device, the models are integrated on the RPi system. Hence the technical combination of Deep Learning
object detection and Internet of Things gives us an opportunity to save lives.
3. BACKGROUND AND PROBLEM WITH EXISTING ART
Before we actually dive into the working snake detector, we should start with the comparison of existing ideas.
4. SUMMARY OF THE INVENTION
Contributing to the global aim of reducing snakebite envenoming the proposed gadget, a unique combination of Deep Learning Object detection frameworks and IoT system is developed to create an impact in rural India. We use the YOLO detection algorithm that has been proved to outperform the previous sliding window object detection. An independent functional system of RPi loaded with trained models which identify onspot live video will alert the surroundings about the danger.
LIST OF PREFERRED AND OPTIONAL FEATURES
1. Linking with coordinates and mapping the exact location of the snake and sending the information to the concerned
wild animaf protection cell.
2. Supplying power to RPI through solar energy is an added benefit for saving electricity.
3. Including better vision cameras which can identify the camouflaged snakes from high resolution cameras and advanced architectures trained on suitable dataset.
4. Alerting based on the level of venomous snake and identifying which snake it is can be very precise and will be an advanced system.
BRIEF DESCRIPTION OF THE DRAWING
Figure: 1 Shows the workflow diagram of the entire setup.
Figure:2 The results obtained during the training phase using YOLO v5 small are promising for just 30 epochs.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
The proposed Snake detector device comprises a Raspberry Pi 4 microcontroller and a night vision camera. For
alerting mechanisms, a red LED via a relay module and a buzzer is used. A program running on the Raspberry Pi constantly reads input feeds from the camera and passes it through the YOLOv5 object detection model to detect the presence of a snake. If a snake is found in the given image, the LED glows and the buzzer rings, alerting the people nearby.
Documents
Application Documents
| # | Name | Date |
|---|---|---|
| 1 | 202241000131-Small Entity_Form-28_03-01-2021.pdf | 2021-01-03 |
| 2 | 202241000131-Educational Institution Eligibility Document_03-12-2021.pdf | 2021-12-03 |
| 3 | 202241000131-Form9_Early Publication_03-01-2022.pdf | 2022-01-03 |
| 4 | 202241000131-Form18_Examination Request_03-01-2022.pdf | 2022-01-03 |
| 5 | 202241000131-Form-5_As Filed_03-01-2022.pdf | 2022-01-03 |
| 6 | 202241000131-Form-3_As Filed_03-01-2022.pdf | 2022-01-03 |
| 7 | 202241000131-Form-1_As Filed_03-01-2022.pdf | 2022-01-03 |
| 8 | 202241000131-Form 2(Title Page)_Complete_03-01-2022.pdf | 2022-01-03 |
| 9 | 202241000131-Drawings_As Filed_03-01-2022.pdf | 2022-01-03 |
| 10 | 202241000131-Description Complete_As Filed_03-01-2022.pdf | 2022-01-03 |
| 11 | 202241000131-Correspondence_As Filed_03-01-2022.pdf | 2022-01-03 |
| 12 | 202241000131-Claims_As Filed_03-01-2022.pdf | 2022-01-03 |
| 13 | 202241000131-Abstract_As Filed_03-01-2022.pdf | 2022-01-03 |
| 14 | 202241000131-FER.pdf | 2022-08-10 |
Search Strategy
| 1 | SearchHistory(5)E_08-08-2022.pdf |