Refine
Document Type
Language
- English (6)
Has Fulltext
- no (6)
Publication reviewed
- begutachtet (6)
Institute
Infrastructure-cooperated autonomous driving systems are attracting attention as a method for promoting the practical application of highly functional autonomous driving. We focused on the part where recognition processing on the infrastructure side can be advanced, which is not possible with in-vehicle processing. Using a fixed-point camera and recognition of the situation behind a vehicle or object in which multiple cameras are linked are such examples. In this paper, we selected a difficult situation such as a curved road, focused on the scene where the vehicle is running while deforming its shape, and examined a method of accurately recognizing the vehicle using a fixed-point camera. It is a study of the criteria for dividing the vehicle shape class. Recognition of the general vehicle class of autonomous driving also needs to identify unknown objects and non-vehicles. In his article, we have excluded the identification of unknown objects and focused on recognizing known vehicles using deep learning. Consider six different vehicle shapes on curved roads. We investigated the impact of vehicle shape class integration and performance, and found that the integration of the two classes reduced the number of vehicle shape classes and increased recognition accuracy.
In autonomous driving, detecting vehicles together with their parts, such as a license plate is important. Many methods with using deep learning detect the license plate based on number recognition. However, there is an idea that the method using deep learning is difficult to use for autonomous driving because of the complexity in realizing deterministic verification. Therefore, development of a method that does not use deep learning(DL) has become important again. Although the authors have made the world's best performance in 2018 for Caltech data with using DL, this concept has now turned to another research without using DL. The CT5L method is the latest type, that includes techniques of the continuity of vertical and horizontal black-and-white pixel values inside the plate, unique Hough transform, only vertical and horizontal lines are detected, the top five in the order of the number of votes to ensure good performance. In this paper, a method to determine the threshold value for binarizing input by machine learning is proposed, and good results are obtained. The detection rate is improved by about 20 points in percent as compared to the fixed case. It achieves the best performance among the conventional fixed threshold method, Otsu's method, and the conventional method of JavaANPR.
There is a method for deterministically detecting a license plate, which is one of the object detections in autonomous driving. It is difficult to determine a luminance threshold value used for binarization. Therefore, the method of predicting the threshold value by machine learning from the combination of surrounding luminances has been improved. First, we constructed an augmentation that extends the objective variable from the structure of the labeled data, and increased the number of the original data by about 50 times. Next, we devised a new method to prevent overestimation of the cross-validation method. After Augmentation, we developed a Leave A Group Out Cross-Validation method that separates training data and test data in groups. By combining the performance improvement by the configured Augmentation and the improvement by the conventional SMOTE, a detection rate of 0.92 was achieved. As a result, the autonomous driving module has been strengthened by one step.
In deep learning, in order to improve learning performance, preprocessing and ingenuity to combine a plurality of discriminators are performed. It can be inferred that it has elements exceeding the set of learning. Therefore, a configuration to combine multiple recognition elements with low loss will be studied. The advance category classification method is expected to narrow the scope of learning in the next stage. Combining elements specialized for FalsePositive/FalseNegative removal after the positive/negative determination is considered to be effective if the accuracy of the subsequent stage is high. We conducted a license plate recognition experiment by combining these and achieved the best performance for Caltech data.
To avoid rear end collisions by following drivers, a high-speed and reliable vehicle detection is needed. One of elements of detecting vehicles is a recognition method of number plates. To detect number plate region, horizontal and vertical differential filters have been used. To improve the precision, we propose a combination of luminance decision and an extended sobel filter.
The Design of a new car moves from real car prototype to software based virtual design methods, see e.g. [1]. These methods accelerate both the design and the development of new functions. Especially safety functions are of interest. The rear-end collision, which is typically caused by other drivers, is an accident that cannot be easily controlled by the ego driver. In the case, a rear-end collision takes place, a fast detection of the status and a fast evacuation are needed. This paper classifies the rear-end collisions and proposes methods of avoiding the accident and if this is not possible of minimizing the damage. Rigorous and accurate evaluation of the risk of the coming collision by fast and real-time processing and reliable detection of accidents should be realized. To improve the reliability, the integration of several detecting elements is needed. At the detection of the collision, it is necessary to minimize false positive and permit false negative identifications. Then, logical AND of results of detection elements is used to exclude the false positive error.
The types of the ego-car positions in rear-end collisions are running and stopping in a traffic jam. Types of driver’s behaviours in following vehicles are e.g. look away, inattentive, careless, dozing, etc. Types of situations are e.g. poor visibility in snow, in the fog, insufficient vehicle distance, etc. The collision should be detected at least two seconds or 50 meters before the accident at the difference speed of 100 km/h. The evacuation may be indicated to the following car by light and sound signals or electric impulses. To judge that the collision shall surely occur, the moving object must be identified as a real vehicle. Detection elements to identify real vehicle are the number plate, vehicle symmetry, the tires, the windscreen, the face or the eyes of the driver etc. In this paper, among the above elements, the number plate detection is deeply improved. The method presented is mainly developed to improve correct detection rate in comparison with conventional methods. The features of the number plate are many vertical and horizontal lines combined to numbers with horizontal to vertical aspect ratio in the order of the detected rectangle of the number plate. Additionally there is a relatively high luminance of background of number plate area. In this paper, in addition to these features, an advanced sobel filter is introduced to adapt to the size variation of the number plate depending on the distance between two cars. The basic coefficients of the original sobel filter is (1,0,-1). The proposed advanced sobel filter is (1,0,0,-1). By using the original and the advanced sobel filters, larger plates and smaller plates will be detected adaptively. The resulted rear-end collision avoidance software will be integrated and tested in the virtual system design approach.