Li-Chih Chen

Symmetrical SURF and Its Applications to Vehicle Detection and Vehicle Make and Model Recognition

Speeded-Up Robust Features (SURF) is a robust and useful feature detector for various vision-based applications but it is unable to detect symmetrical objects. This paper proposes a new symmetrical SURF descriptor to enrich the power of SURF to detect all possible symmetrical matching pairs through a mirroring transformation. A vehicle make and model recognition (MMR) application is then adopted to prove the practicability and feasibility of the method. To detect vehicles from the road, the proposed symmetrical descriptor is first applied to determine the region of interest of each vehicle from the road without using any motion features. This scheme provides two advantages: there is no need for background subtraction and it is extremely efficient for real-time applications. Two MMR challenges, namely multiplicity and ambiguity problems, are then addressed. The multiplicity problem stems from one vehicle model often having different model shapes on the road. The ambiguity problem results from vehicles from different companies often sharing similar shapes. To address these two problems, a grid division scheme is proposed to separate a vehicle into several grids; different weak classifiers that are trained on these grids are then integrated to build a strong ensemble classifier. The histogram of gradient and SURF descriptors are adopted to train the weak classifiers through a support vector machine learning algorithm. Because of the rich representation power of the grid-based method and the high accuracy of vehicle detection, the ensemble classifier can accurately recognize each vehicle.

Vehicle Color Classification Under Different Lighting Conditions Through Color Correction

This paper presents a novel color correction technique for classifying vehicles under different lighting conditions using their colors. To reduce the lighting effects, a reference image is first selected for building the mapping function between the current frame and the reference image. With this mapping function, the color distortions between frames can be reduced to minimum. In addition to lighting changes, the effect of sun light will make the vehicle window become white and lead to the errors of vehicle classification. To reduce this effect, a window-removing task is then applied for making vehicle pixels with the same color more concentrated on the foreground region. Then, vehicles can be more accurately classified to their categories even though strong sun light casts on them. To tackle the confusion problem that some vehicle colors are too similar, e.g., “deep-blue” and “deepgreen”, a novel tree-based classifier is then designed for classifying vehicles to more detailed labels. Experimental results have proved that the proposed method is a robust, accurate, and powerful tool for vehicle classification.

Vision-Based Vehicle Surveillance and Parking Lot Management Using Multiple Cameras

This paper proposes a vision-based vehicle surveillance system for parking lot management in outdoor environments. Due to the limited field of view of camera, this system uses multiple cameras for monitoring a wide parking area. Then, an affine transformation is used for merging the scenes obtained from these multiple cameras. Two major components are included, i.e., vehicle counting and parking lot management. For the first one, this paper integrates three features, i.e., color, position, and motion together for well tracking vehicles across different cameras. Thus, even though vehicles are occluded together, they still can be well tracked and identified across different cameras and under different lighting changes. For the second one, we propose a model-based approach to model the color changes of parking ground for determining whether a parking space is vacant. Due to the perspective effects, the visibility of a parking space is often affected by the vehicle parking on its neighborhood. To tackle this problem, two geometrical models (ellipses and grids) are proposed for well representing a parking space. Then, with different weights, a hybrid scheme is then constructed for well determining whether a parking space is vacant. The experimental results reveal that our system works well and accurately under different lighting and occlusion conditions.

Nighttime Turn Signal Detection by Scatter Modeling and Reflectance-Based Direction Recognition

The rapid expansion of car ownership worldwide has further raised the importance of vehicle safety. The reduced cost of cameras and optical devices has made it economically feasible to deploy front-mounted intelligent systems for visual-based event detection for forward collision avoidance and mitigation. While driving at night, vehicles in front are generally visible by their tail lights. The turn signals are particularly important because they signal lane change and potential collision. Therefore, this paper proposes a novel visual-based approach, based on the Nakagami-m distribution, for detecting turn signals at night by scatter modeling of tail lights. In addition, to recognize the direction of turn signals, reflectance is decomposed from the original image. Rather than using knowledge of heuristic features, such as the symmetry, position, and size of the rear-facing vehicle, we focus on finding the invariant features to model turn signal scattering by Nakagami imaging and therefore, conduct the detection process in a part-based manner. Experiments on an extensive data set show that our proposed system can effectively detect vehicle braking under different lighting and traffic conditions, and thus, demonstrates its feasibility in real-world environments.

Vehicle make and model recognition using symmetrical SURF

SURF (Speeded Up Robust Features) is a robust and useful feature detector for various vision-based applications but lacks the ability to detect symmetrical objects. This paper proposes a new symmetrical SURF descriptor to enrich the power of SURF to detect all possible symmetrical matching pairs through a mirroring transformation. A vehicle make-and-model recognition (MMR) application is then adopted to prove the practicability and feasibility of the method. To detect vehicles from the road, the proposed symmetrical descriptor is first applied to determine the ROI of each vehicle from the road without using any motion features. This scheme provides two advantages; there is no need of background subtraction and it is extremely efficient for real-time applications. Two MMR challenges, i.e., multiplicity and ambiguity problems, are then addressed. The multiplicity problem stems from one vehicle model often having different model shapes on the road. The ambiguity problem results from vehicles from different companies often sharing similar shapes. To address these two problems, a grid division scheme is proposed to separate a vehicle into several grids; different weak classifiers that are trained on these grids are then integrated to build a strong ensemble classifier. Because of the rich representation power of the grid-based method and the high accuracy of vehicle detection, the ensemble classifier can accurately recognize each vehicle.

Vehicle make and model recognition using sparse representation and symmetrical SURFs

This paper proposes a new symmetrical SURF descriptor to detect vehicles on roads and applies the sparse representation for the application of vehicle make-and-model recognition (MMR). To detect vehicles from roads, this paper proposes a symmetry transformation on SURF points to detect all possible matching pairs of symmetrical SURF points. Then, each desired ROI of vehicle can be located very accurately through a projection technique. This scheme provides two advantages; there is no need of background subtraction and it is extremely efficient for real-time applications. Two MMR challenges, i.e., multiplicity and ambiguity problems, are then addressed. The multiplicity problem stems from one vehicle model often having different model shapes on the road. The ambiguity problem results from vehicles from different companies often sharing similar shapes. To treat the two problems, a dynamic sparse representation scheme is proposed to represent a vehicle model in an over-complete dictionary whose base elements are the training samples themselves. With the dictionary, a novel Hamming distance classification scheme is proposed to classify vehicle makes and models to detailed classes. Because of the sparsity of sparse representation and the nature of Hamming code highly tolerant to noise, different vehicle makes and models can be recognized extreme accurately.

Sparse representation for recognizing object-to-object actions under occlusions

In this paper, we describe the formatting guidelines for ACM SIG Proceedings. This paper proposes a novel event classification scheme to analyze various interaction actions between persons using sparse representation. The occlusion problem and the high complexity to model complicated interactions are two major challenges in person-to-person action analysis. To address the occlusion problem, the proposed scheme represents an action sample in an over-complete dictionary whose base elements are the training samples themselves. This representation is naturally sparse and makes errors (caused by different environmental changes like lighting or occlusions) sparsely appear in the training library. Because of the sparsity, it is robust to occlusions and lighting changes. In addition, a novel Hamming distance classification (HDC) scheme is proposed to classify action events to detailed types. Because the nature of Hamming code is highly tolerant to noise, the HDC scheme is also robust to occlusions. The high complexity of complicated action modeling can be tackled by adding more examples to the over-complete dictionary. Thus, even though the interaction relations are complicated, the proposed method still works successfully to recognize them and can be easily extended to analyze action events among multiple persons. More importantly, the HDC scheme is very efficient and suitable for real-time applications because no optimization process is involved to calculate the reconstruction error.

Real-time vehicle color identification using symmetrical SURFs and chromatic strength

This paper proposes a new vehicle color classification scheme to identify vehicles with their colors. To detect vehicles from roads, the paper proposes a novel symmetrical descriptor to determine the ROI of each vehicle without using any motion features. This scheme provides two advantages; there is no need of background subtraction and it is extremely efficient for real-time applications. After detection, a novel color-correction technique is proposed to reduce the color changes of vehicles so that vehicles can be more accurately identified. The major challenge in vehicle color identification is there are many shade (or confused) colors among vehicles. This paper proposes a new concept that the vehicles with different chromatic attributes should be separately trained even though they are in the same color category. With this concept, a novel tree-based classifier can be constructed to classify vehicles at different stages according to their chromatic strengths. The separation can significantly improve the accuracy of vehicle color classification even that vehicles are with various shade colors.

Robust Rear Light Status Recognition Using Symmetrical SURFs

This paper proposes a new framework to detect vehicle indicator lights and recognize their statuses using symmetrical SRUFs. To detect indicator lights from a vehicle, a symmetrical descriptor is first applied to determine its position from roads. Two advantages can be gained from this scheme, there is no need of background subtraction and it is extremely efficient for real-time analysis applications. After vehicle detection, a new lamp response function is defined for isolating red components from the detected vehicle for rear lamp detection without using any thresholds. This is very different and superior to other state-of-art frameworks in the literature. The positions of rear lamp can be then accurately located by searching the peaks of lamp response function even under daytime or nighttime conditions. To recognize the statuses of a rear lamp, no training stage is needed to train a classifier for lamp status analysis. To achieve this goal, a new mask is designed to make status judgments on a lamp according to only its response sign. Because no any threshold is adopted, various rear lamps and their statuses can be accurately analyzed even under various lighting conditions.

PLSA-based sparse representation for vehicle color classification

This paper proposes a novel vehicle color classification method which uses the concept of probabilistic latent semantic analysis (pLSA) to overcome the problem of sparse representation in data classification. Sparse representation is widely used and quite successful in many vision-based applications. However, it needs to calculate the sparse reconstruction cost (SRC) of each sample to find the best candidate. Because an optimization process is involved, it is very inefficient. In addition, it uses only the residual and does not consider the arrangement (or distribution) of combination coefficients of visual codes in classification. Thus, it often fails to classify categories if they are similar. In this paper, the pLSA concept is first introduced into the sparse representation to build a new classifier without using the SRC measure. The weakness of the pLSA scheme is the use of EM algorithm for updating the posteriori probability of latent class. Because it is very time-consuming, a novel weighting voting strategy is introduced to improve the pLSA scheme for recognizing objects in real time. The advantages of this classifier are: the accuracy is much higher than the SRC scheme and the efficiency is real-time in data classification. Vehicle color classification is demonstrated in this paper to prove the superiority of the new classifier.