Detection, Tracking and Classification of Road Signs in Adverse Conditions

Author(s):  
G.K. Siogkas ◽  
E.S. Dermatas
2012 ◽  
Vol 19 (3) ◽  
pp. 285-298 ◽  
Author(s):  
Gustavo A. Peláez Coronado ◽  
María Romero Muñoz ◽  
José María Armingol ◽  
Arturo de la Escalera ◽  
Juan Jesús Muñoz ◽  
...  

Author(s):  
A. Adam ◽  
C. Ioannidis

This paper examines the detection and classification of road signs in color-images acquired by a low cost camera mounted on a moving vehicle. A new method for the detection and classification of road signs is proposed based on color based detection, in order to locate regions of interest. Then, a circular Hough transform is applied to complete detection taking advantage of the shape properties of the road signs. The regions of interest are finally represented using HOG descriptors and are fed into trained Support Vector Machines (SVMs) in order to be recognized. For the training procedure, a database with several training examples depicting Greek road sings has been developed. Many experiments have been conducted and are presented, to measure the efficiency of the proposed methodology especially under adverse weather conditions and poor illumination. For the experiments training datasets consisting of different number of examples were used and the results are presented, along with some possible extensions of this work.


Author(s):  
Amal Bouti ◽  
Mohamed Adnane Mahraz ◽  
Jamal Riffi ◽  
Hamid Tairi

In this chapter, the authors report a system for detection and classification of road signs. This system consists of two parts. The first part detects the road signs in real time. The second part classifies the German traffic signs (GTSRB) dataset and makes the prediction using the road signs detected in the first part to test the effectiveness. The authors used HOG and SVM in the detection part to detect the road signs captured by the camera. Then they used a convolutional neural network based on the LeNet model in which some modifications were added in the classification part. The system obtains an accuracy rate of 96.85% in the detection part and 96.23% in the classification part.


2019 ◽  
Vol 8 (1) ◽  
pp. 65-69
Author(s):  
Tatyana Borisovna Matveeva ◽  
Ivan Victorovich Kazantsev ◽  
Sergey Lvovich Molchatsky

The paper presents data on the state of soil and vegetation cover of oak forests of suburban forests near the village Mekhzavod. Being in the ring of large highways as well as located relatively close to Samara, they experience a complex impact of many adverse conditions. In the course of the study in these forest communities using the method of laying ecological profiles, we assessed a degree of recreational load impact on the vegetation cover of the green zone. The author estimated the percentage of road and path network development, described stages of recreational digression (for R.A. Karpisonova) and identified the main indicators of anthropogenic load, a vital status of major forest tree species and the factors contributing to its deterioration. The author also revealed a classification of soils with the help of laying soil profiles in different quarters and the subsequent chemical analysis of the selected samples. It is determined that the gray forest soils indicated for this area on the soil map of the Volga Region are not found. Taking into account the unsatisfactory state of the vegetation cover of the studied area, a number of recommendations for its rational use are proposed, which can further contribute to the preservation and improvement of these forests stability.


2007 ◽  
Vol 17 (3) ◽  
pp. 265-289 ◽  
Author(s):  
Yok-Yen Nguwi ◽  
Abbas Z. Kouzani

Author(s):  
А.О. Калашников ◽  
В.Ф. Барабанов ◽  
А.М. Нужный ◽  
А.В. Барабанов

Рассмотрены вопросы создания системы поддержки принятия решений при составлении паспорта дороги. Одной из задач, решаемых в процессе паспортизации автомобильной дороги, является заполнение данных о наличии и расположении искусственных сооружений, дорожных инженерных устройств, в том числе дорожных знаков. Система предназначена для автоматизированного анализа видеопотока с целью выделения кадров, содержащих дорожные знаки, а также последующей классификации найденных знаков. Приведена оригинальная двухэтапная система извлечения и классификации изображений, содержащих дорожные знаки. Алгоритм обнаружения требуемых изображений базируется на использовании вейвлет-преобразований Хаара и концепции интегрального изображения, что позволяет максимально быстро находить требуемые кадры. Оригинальность применения признаков Хаара состоит в том, чтобы использовать только 2 прямоугольных фильтра (горизонтальный и вертикальный) в разных масштабах: 2x2, 4x4, 8x8 и 16x16. Последующая обработка данных, целью которой является классификация найденных изображений, осуществляется с применением искусственной нейронной сети. Актуальность разработки подобной системы поддержки принятия решения определяется необходимостью обработки больших объемов видеоданных. Система позволяет в значительной мере исключить фактор пользовательских ошибок, что очень важно, так как полученные данные влияют на безопасность дорожного движения The article considers issues of creating a decision support system for the preparation of the passport of a road. One of the tasks solved in the process of passporting a motorway is to fill data on the presence and location of structures, road engineering devices, including road signs. The system is designed for automated analysis of the video flow to highlight frames containing road signs, as well as the subsequent classification of the found characters. We give an original two-stage system for extracting and classifying images containing road signs. The detection algorithm for the desired images is based on the use of Haar's wavelet transforms and the concept of an integrated image, which allows one to get the required frames as quickly as possible. The originality of the use of Haar signs is to use only 2 rectangular filters (horizontal and vertical) on different scales: 2x2, 4x4, 8x8 and 16x16. Subsequent data processing, the purpose of which is the classification of the found images, is carried out using an artificial neural network. The relevance of the development of such a decision support system is determined by the need to process large volumes of video data. The system allows one to largely eliminate the factor of user errors, which is very important since the data obtained affect the safety of the road


Sign in / Sign up

Export Citation Format

Share Document