A Cognitively Motivated Method for Classification of Occluded Traffic Signs

A Cognitively Motivated Method for Classification of Occluded Traffic Signs

ABSTRACT:

Classification of traffic signs with partial occlusions is important for traffic sign maintenance and inventory systems. It is also important to help drivers identify possible traffic signs in time. Motivated by human cognitive processes in identifying an occluded sign, a novel structure is designed to explicitly handle occluded samples in this paper. Occlusion maps are analyzed for possible occluded signs, and a new occlusion descriptor is proposed to distinguish occluded signs from negative samples. A series of tests shows that the developed method could effectively handle samples with partial occlusions and thus reduce the missed detections caused by occlusions. The developed method could also be easily used for any other object detection.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• They applied Convolutional Networks (ConvNets) to the task of traffic sign classification as part of the GTSRB competition. ConvNets are biologically-inspired multi-stage architectures that automatically learn hierarchies of invariant features.

• While many popular vision approaches use handcrafted features such as HOG or SIFT, ConvNets learn features at every level from data that are tuned to the task at hand.

• The traditional ConvNet architecture was modified by feeding 1 st stage features in addition to 2 nd stage features to the classifier.

DISADVANTAGES OF EXISTING SYSTEM:

• Existing method achieved high recognition rates in the competitions held by IJCNN. However, the real-time application of these methods is constrained by the memory and processing ability of the current embedded systems.

• To the best of our knowledge, until now, most methods have been based on holistic features in a candidate region. Their ability to deal with signs with partial occlusions mainly relies on a huge amount of training data. These methods may easily fail to detect signs partially occluded by objects in the streets due to the insufficient visual clues.

PROPOSED SYSTEM:

• In this paper, to deal with occluded samples in the HOG-based method, a novel structure is designed for occlusion analysis in the first stage. The idea is motivated by the human cognitive process for recognizing an occluded sign.

• First, knowledge about the object of interest is gathered based on the number of entire objects.

• Second, a visibility map is established for a given image based on learned knowledge.

• Finally, if a part of an object can be vividly seen but the other parts are not visible, the image is usually viewed as an occluded version of the object of interest.

• The newly designed implementation of the first stage showed. During training, knowledge about traffic signs in each category is learned based on a number of training samples.

• During testing, samples are classified into each category. Samples classified with a high confidence are passed to the second stage directly. The occlusion maps of samples with a low confidence are analyzed. An occlusion feature is developed to describe the occlusion maps, and the classification decision will be made based on the occlusion descriptor.

ADVANTAGES OF PROPOSED SYSTEM:

• The results show that the new method can effectively distinguish occluded samples from negative samples, thus reducing missed detections caused by partial occlusions.

• No significant extra processing time would be necessary for each frame.

MODULES:

1. Image Acquisition

2. Occluded Traffic Sign Detection

3. Feature Extraction

4. Classification

MODULE DESCRIPTION:

1. Image Acquisition:

• Image acquisition in image processing can be broadly defined as the action of retrieving an image from some source, usually a hardware-based source, so it can be passed through whatever processes need to occur afterward.

• Performing image acquisition in image processing is always the first step in the workflow sequence because, without an image, no processing is possible. The image that is acquired is completely unprocessed and is the result of whatever hardware was used to generate it, which can be very important in some fields to have a consistent baseline from which to work.

• Images are acquired from Gallery.

2. Occluded Traffic Sign Detection:

• This is the second stage of our work. In that occluded Traffic sign was detected by high intensity pixels.

• First median filter was applied for remove the noise in traffic sign.

• After that high intensity pixel or red color detection based traffic sign was detected.

3. Feature Extraction:

• After traffic sign detection, Feature extraction is performed. It is done by HOG feature descriptor.

• HOG descriptor is mainly suitable for traffic sign detection in video or images due to some key advantages compared to other descriptors. First, it operates on local cells, so it is invariant to geometric and photometric transformations. Secondly, coarse (spatial) sampling, fine orientation sampling, and strong local photometric normalization. After that extracted features are given to input of classification.

4. Classification:

• The classification process is done over the extracted features. The main novelty here is the adoption of multiclass SVM. SVM classifier is applied over the features and the classification is done.

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 7/10/11.

• Coding Language : MATLAB

• Tool                       : MATLAB R2013A

REFERENCE:

Ya-Li Hou, Xiaoli Hao, and Houjin Chen, “A Cognitively Motivated Method for Classification of Occluded Traffic Signs”, IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS: SYSTEMS, 2017.