• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.663-672
• /
• 2023

Most vehicle detection methods have poor vehicle feature extraction performance at night, and their robustness is reduced; hence, this study proposes a night vehicle detection method based on style transfer image enhancement. First, a style transfer model is constructed using cycle generative adversarial networks (cycleGANs). The daytime data in the BDD100K dataset were converted into nighttime data to form a style dataset. The dataset was then divided using its labels. Finally, based on a YOLOv5s network, a nighttime vehicle image is detected for the reliable recognition of vehicle information in a complex environment. The experimental results of the proposed method based on the BDD100K dataset show that the transferred night vehicle images are clear and meet the requirements. The precision, recall, mAP@.5, and mAP@.5:.95 reached 0.696, 0.292, 0.761, and 0.454, respectively.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2017년도 제56차 하계학술대회논문집 25권2호
• /
• pp.135-136
• /
• 2017

When driving at night time environment, the whole body of transports does not visible to us. Due to lack of light conditions, there are only two options, which is clearly visible their taillights and break lights. To improve the recognition correctness of vehicle detection, we present an approach to vehicle detection and tracking using finding contour of the object on binary image at night time. Bilateral filtering is used to make more clearly on threshold part. To remove unexpected small noises used morphological opening. In verification stage, paired tail lights are tracked during their existence in the ROI. The accuracy of the test results for vehicle detection is about 93%.

• 대한임베디드공학회논문지
• /
• 제7권4호
• /
• pp.163-172
• /
• 2012

A driver assistance system is critical to improve a convenience and stability of vehicle driving. Several systems have been already commercialized such as adaptive cruise control system and forward collision warning system. Efficient vehicle detection is very important to improve such driver assistance systems. Most existing vehicle detection systems are based on a radar system, which measures distance between a host and leading (or oncoming) vehicles under various weather conditions. However, it requires high deployment cost and complexity overload when there are many vehicles. A camera based vehicle detection technique is also good alternative method because of low cost and simple implementation. In general, night time vehicle detection is more complicated than day time vehicle detection, because it is much more difficult to distinguish the vehicle's features such as outline and color under the dim environment. This paper proposes a method to detect vehicles at night time using analysis of a captured color space with reduction of reflection and other light sources in images. Four colors spaces, namely RGB, YCbCr, normalized RGB and Ruta-RGB, are compared each other and evaluated. A suboptimal threshold value is determined by Otsu algorithm and applied to extract candidates of taillights of leading vehicles. Statistical features such as mean, variance, skewness, kurtosis, and entropy are extracted from the candidate regions and used as feature vector for SVM(Support Vector Machine) classifier. According to our simulation results, the proposed statistical feature based SVM provides relatively high performances of leading vehicle detection with various distances in variable nighttime environments.

• 대한임베디드공학회논문지
• /
• 제6권5호
• /
• pp.311-317
• /
• 2011

The blind spot is an area where drivers visibility does not reach. When drivers change a lane to adjacent lane, they need to give an attention because of the blind spot. If drivers try to change lane without notice of vehicle approaching in the blind spot, it causes a reason to have a car accident. This paper proposes a night-time blind spot vehicle detection using cameras. At nighttime, head-lights are used as characteristics to detect vehicles. Candidates of headlight are selected by high luminance feature and then shape filter and kalman filter are employed to remove other noisy blobs having similar luminance to head-lights. In addition, vehicle position is estimated from detected head-light, using virtual center line represented by approximated the first order linear equation. Experiments show that proposed method has relatively high detection porformance in clear weather independent to the road types, but has not sufficient performance in rainy weather because of various ground reflectors.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권4호
• /
• pp.269-275
• /
• 2017

Automatic vehicle detection is a very important function for autonomous vehicles. Conventional vehicle detection approaches are based on visible-light images obtained from cameras mounted on a vehicle in the daytime. However, unlike daytime, a visible-light image is generally dark at night, and the contrast is low, which makes it difficult to recognize a vehicle. As a feature point that can be used even in the low light conditions of nighttime, the rear lamp is virtually unique. However, conventional rear lamp-based detection methods seldom cope with atypical lamps, such as LED lamps, or flashing turn signals. In this paper, we detect atypical lamps by blurring the lamp area with a low pass filter (LPF) to make out the lamp shape. We also propose to detect flickering of the turn signal lamp in a manner such that the lamp area is vertically projected, and the maximum difference of two paired lamps is examined. Experimental results show that the proposed algorithm has a higher F-measure value of 0.24 than the conventional lamp pairing-based detection methods, on average. In addition, the proposed algorithm shows a fast processing time of 6.4 ms per frame, which verifies real-time performance of the proposed algorithm.

• 한국정보전자통신기술학회논문지
• /
• 제8권4호
• /
• pp.270-274
• /
• 2015

본 논문에서는 HSV 색 공간을 이용한 야간 차량의 검출 시스템을 제안한다. 주정차 감시등 도로변에서 자동차를 감시하는 경우 자동차 번호판 추출하는 것이 중요하다. 일반적으로 번호판 추출을 위해서는 원거리에서 자동차 검출후 Pan-Tilt-Zoom 카메라로 자동차를 일정한 크기로 확대한 영상을 획득하여 번호판을 추출한다. 그리고 자동차 검출 및 추적을 위해 Mean-Shift 혹은 Optical Flow 알고리듬이 많이 이용되고 있다. 그러나 이러한 알고리즘은 주간에는 성공적으로 자동차를 검출 및 추적 할수 있었으나 야간에는 검출 및 추적에 어려움이 있었다. 그래서 본 논문에서는 입력 영상을 HSV 색 공간으로 변환하면 자동차의 전조등 혹은 후미등의 위치가 두드러지게 나타나는 것을 이용하여 자동차의 위치를 검출하였다. 실험 결과 정면 차량의 경우 93.9%, 후면 차량의 경우 97.7%의 차량을 검출하여 제안된 방법이 야간 차량 검출에 효율적임을 증명하였다.

• 정보처리학회논문지B
• /
• 제18B권4호
• /
• pp.201-212
• /
• 2011

본 논문에서는 특징점 추적을 이용한 끼어들기 위반차량 검지 시스템을 제안한다. 제안된 끼어들기 위반차량 검지 시스템의 전체적인 알고리즘은 영상 변환 및 전처리, 특징 추출, 추적대상 차량의 특징점 등록 및 추적, 끼어들기 위반차량 검지 등의 단계로 구성된다. 특히 형태학적 기울기 영상을 이용하여 특징점을 추출하므로 써 주간 및 야간 영상에 대해 동일한 알고리즘을 적용하여 그림자, 기상 조건, 차량 전조등 및 조명 등에 강인한 영상 검지 시스템을 구성 한다. 제안한 시스템을 끼어들기 금지구간에서 주간, 야간, 비 오는 날 야간에 취득한 영상을 사용하여 실험한 결과 정인식률 99.49%와 오류율 0.51%를 보였다. 또한 실시간처리에 문제가 없는 평균 91.34frame/s의 빠른 처리속도를 나타냈다.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1995년도 춘계공동학술대회논문집; 전남대학교; 28-29 Apr. 1995
• /
• pp.316-323
• /
• 1995

This paper presents an image processing technique to get traffic information such as vehicle volume, velocity, and occupancy for measuring the traffic congestion rate. To obtain these information, two horizontal lines are previously set on the screen. A moving vehicle is detected using the gray level difference on each line, and also template matching method at night. Threshold values are determined by sampling pavement grey level, and updated dynamically to cope with the change of ambient light conditions. These technique is successfully used to calculate vehicle volume, occupancy, and velocity. This study can be applied to traffic signal control system for minimizing traffic congestion in urban areas.

• 대한임베디드공학회논문지
• /
• 제10권2호
• /
• pp.101-108
• /
• 2015

In this paper, a robust traffic light detection method is proposed by using vision sensor and DGPS(Difference Global Positioning System). The conventional vision-based detection methods are very sensitive to illumination change, for instance, low visibility at night time or highly reflection by bright light. To solve these limitations in visual sensor, DGPS is incorporated to determine the location and shape of traffic lights which are available from traffic light database. Furthermore the geometric relationship between traffic light and vision sensor is used to locate the traffic light in the image by using DGPS information. The empirical results show that the proposed method improves by 51% in detection rate for night time with marginal improvement in daytime environment.

• 전자공학회논문지SC
• /
• 제42권2호
• /
• pp.11-18
• /
• 2005

본 논문은 CCD 카메라를 통해 입력받은 영상에서 차량을 검출하는 방법을 제안한다. 차량을 검출하기 위해서 먼저 영상을 독립적인 방향과 레벨을 가지는 스티어블 피라미드로 변환한다. 특징 벡터는 스티어블 피라미드로 변환된 서브밴드들을 연관되는 같은 위치의 픽셀들을 체인으로 연결하여 방향성 피라미드 특징을 가지는 다차원 벡터들로 구성한다. 차량의 검출은 특징 점의 특징 벡터들을 차량 검출에 사용하였다. 특징 점은 기하학적 위치 정보와 국부적인 방향 정보를 가지는데 실험을 위해서 격자 구조 모양으로 일정한 간격을 갖는 격자 점, 사람의 수작업을 통해서 만든 코너 점, 그리고 격자 내의 코너 점을 대상으로 했다. 차량 검출을 위해 미리 저장된 모델 영상의 특징 점들의 특징벡터들과 후보 영상으로부터 추출된 특징 벡터들의 정합을 통해 각 특징 점의 거리를 비교했다. 차량 검출을 위해 특징 점을 이용함으로써 후보 영상 전체를 비교하지 않고 특징 점의 위치에 대해서만 특징 벡터를 비교하기 때문에 비교 시간과 정확도를 높일 수 있었다. 또한 주변 밝기조건 및 그림자의 영향에 의해 차량 검출이 민감한 문제를 해결할 수 있었다. 도로에서 획득한 주간 영상(10,567)과 저녁 영상(624)을 대상으로 실험하였고, 검출율은 주간의 경우 $92.0\%$와 야간의 경우 $87.3\%$를 얻을 수 있었다.