Night Time Vehicle detection and tracking based on SVM

ABSTRACT

During day time numerous approaches have been implemented on vehicle detection and tracking. It is very much easier to detect vehicles during day time compared to night time. We can capture photo of every part of vehicle during day. But the appearance of vehicles during night time is strikingly different when compared to its daylight counterpart as several attributes come into picture such as surrounding light, color of vehicles, reflection of lights on the body of vehicles, etc. It becomes difficult to detect or take a picture of every parts of vehicle. When driving in dark conditions, we cannot see the whole body of the vehicle present in front of us due to lack of light conditions, we are only able to see their tail lights and brake lights. In this paper, we proposed vehicle detection and tracking based on tail lights segmentation. We followed five steps for vehicle detection and tracking. The steps are video acquisition, preprocessing, segmentation, feature extraction and classification. We demonstrated our method gives better result than state of arts.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

The existing paper describes a system for detecting vehicles based on their rear lights. This system focuses on close range detection so that the distance in the image between the rear lights is large enough so that the individual lights are distinguishable. As the target vehicle gets further away, the rear lights tend to blur together, resulting in the distortion of the distinctive characteristics used for detection. Of course, the near range is also the area which is most critical in collision detection systems.

DISADVANTAGES:

1. They do not account for vehicles that do not meet legislative requirements, such as vehicles with broken lights or modified lights that do not meet the common legal specification of color, brightness and position.

2. The detection accuracy is less.

PROPOSED SYSTEM:

The proposed system has following steps,

1. Nighttime traffic videos are gathered from public database.

2. The second step is preprocessing, in that acquired video is processed into frame conversion.

3. The third step is segmentation, Thresholding method is used for initial segmentation after that morphological operation is explored for small objects removal, holes filling and connects the small holes into big object in segmented frame. In that stage, we obtained segmentation of tail lights.

4. The fourth stage is feature extraction, the histogram of oriented gradients (HOG) is a feature descriptor used in relevant features extraction.

5. The final stage is classification, vehicle detection and tracking is done by using Support Vector Machine (SVM). The extracted relevant features are given to input of SVM.

ADVANTAGES:

1. The rate of detection accuracy is high.

2. Execution time is less.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB

• Tool                       : MATLAB R2013A