In-vehicle Camera Images Research Articles

ROI based High Efficiency Image Coding with Adaptive Scaling for In-vehicle Camera Image

ROI based High Efficiency Image Coding with Adaptive Scaling for In-vehicle Camera Image

Image-based Visibility Estimation of Road Signs in Cluttered Environment

This paper proposes an imaging-based system to estimate road sign visibility in a cluttered environment from the driver’s perspective in daytime using in-vehicle camera images. The proposed system can be deployed in both Driver Assistance Systems (DAS) and Driver Safety Support Systems (DSSS) as a choosing criterion of what information to provide to drivers. Driver Assistance Systems can only provide drivers with warnings about road signs with less visibility and high importance. The proposed system estimates the visibility by measuring the detect-ability of road signs. The detect-ability parameter measures the ability of the driver to locate road object from a scene and thus; it measures sign postage with respect to cluttered or complex environment. Road signs posted in complex environment are harder to be recognized by drivers and thus have low value for the detect-ability parameter. This paper proposes a visibility estimation system of road signs in the United States and experimental results are used to show its effectiveness.

Estimate the Amount of Disaster Waste Disposal Work Using In-Vehicle Camera Images – A Case Study in Hitoyoshi City, Kumamoto Prefecture –

In recent years, a wide-area disaster causing enormous damage has occurred almost every year in Japan. The authors have been involved in the management of disaster volunteer centers in various places and have realized the difficulty in coordinating the actual number of activities in disaster-stricken areas and the required number of disaster volunteers. The number of disaster volunteers required varies greatly on a daily basis. The number of volunteer activities often depends on the quantity of disaster waste created from damaged houses. Accordingly, if the quantity of waste was grasped immediately on the spot, the number of disaster volunteers required in the short term could be estimated. In this study, the actual conditions in Hitoyoshi City, Kumamoto Prefecture, at the time of the Heavy Rainfall in July 2020 are considered as an example, and the method is considered to immediately grasp the quantity of waste to be disposed of and the number of people required for this task. We compared the amount of waste disposal work estimated from the in-vehicle camera image with the number of active residents and volunteers. As a result, it was estimated that 40–50% of the work was carried out by volunteers at the peak of volunteer activities.

Method for Texture Mapping of In-Vehicle Camera Image in 3D Map Creation

Method for texture mapping of in-vehicle camera image in 3D map creation is proposed. Top view of ground cover targets can be mapped easily. For instance, aerial photos, high spatial resolution of satellite imagery data allows creation of top view of ground cover targets and also map creation. It can be used for pedestrian navigations. On the other hand, side view of ground cover targets is not so easy to obtain. In this paper, two methods are proposed. One is to use acquired photos with cameras mounted on the dedicated cars. The other one is to use high spatial resolution of satellite imagery data, such as IKONOS, Orbview, etc. Through experiments with the aforementioned two methods, it is found that texture mapping for the ground cover targets can be done with the two proposed methods in an efficient manner.

Another View Point Image Generation with Shifted Perspective Using Two In-Vehicle Camera Images for Car Navigation System Based on Vehicular Ad-hoc Networks

It is expected that telepresence techniques are introduced into a car navigation system. For example, the navigation system presents a road scene taken from different points of view to drivers. The realization of the system require cameras on a road demanded by drivers. However, available cameras of roads are generally insufficient. We aim to construct an another view point image generation with shifted perspective using two in-vehicle camera images based on ad-hoc networks. Using the models of a few virtual planes produced from measurement points, we have shown that the transformation is performed in real time. Experimental results demonstrated the effectiveness of the proposed method.

3-5 2枚の車載カメラ画像からの視点変更画像生成に関する検討(第3部門 コンピュータビジョン)

3-5 2枚の車載カメラ画像からの視点変更画像生成に関する検討(第3部門 コンピュータビジョン)

Maximum likelihood estimation of vehicle position for outdoor image sensor-based visible light positioning system

Image sensor-based visible light positioning can be applied not only to indoor environments but also to outdoor environments. To determine the performance bounds of the positioning accuracy from the view of statistical optimization for an outdoor image sensor-based visible light positioning system, we analyze and derive the maximum likelihood estimation and corresponding Cramer–Rao lower bounds of vehicle position, under the condition that the observation values of the light-emitting diode (LED) imaging points are affected by white Gaussian noise. For typical parameters of an LED traffic light and in-vehicle camera image sensor, simulation results show that accurate estimates are available, with positioning error generally less than 0.1 m at a communication distance of 30 m between the LED array transmitter and the camera receiver. With the communication distance being constant, the positioning accuracy depends on the number of LEDs used, the focal length of the lens, the pixel size, and the frame rate of the camera receiver.

誤検出マイニングに基づくシーン適応型歩行者検出

In recent years, demands for Advanced Driving Assistance Systems (ADAS) is increasing, and pedestrian detection has become one of the most important and popular technologies in this system. In the case of pedestrian detection using an in-vehicle camera, since the road environment varies widely according to difference in lightning, weather, etc., it is very difficult to handle them with a single classifier, and numerous false positives are detected. To overcome this problem, this paper proposes a novel pedestrian detection method by scene adaptation based on false positive mining. When we observe the appearance of false positives in in-vehicle camera images, those with similar features are found even in different road environments. The proposed method focuses on the appearance of the detected false positives, and considers it as a scene that the classifier is not good at. By analyzing such a false positive tendency in each scene, the proposed method associates the false positive tendency to each scene and then associates them to each training image. Then, classifiers are constructed so that they can cope with false positives observed in each scene. To evaluate the effectiveness of the proposed method, experiments were conducted on the Caltech Pedestrian Detection Benchmark datasets. Its results showed that the proposed method outperforms the method without adaptation.

拡張Census Transformを用いた道路面の経時差分による車載カメラ映像からの不特定障害物検出

This paper proposes a method for detecting general obstacles on a road by subtracting a present in-vehicle camera image sequence from its corresponding past image sequences using an illumination insensitive feature. We conducted experiments using several image sequences captured under various illumination conditions, and confirmed the effectiveness of the proposed method.

規範フロー場と実フロー場の差異解析による路上障害物の検出と危険度提示に関する研究

Many researches have been done as for outdoor scene recognition by analyzing optical flow information. However, a practical driver-oriented system has not been proposed yet that notices the “Risk degree” when multiple obstacles exist, such as, parking cars and pedestrians. This article presents a new method to detect stand still obstacles using the “model flow” corresponding to optical flow field where there is no obstacle. The model flow is calculated using the knowledge as for navigation environments. Standard Points which are regularly set on image coordinates are transformed to Environment Points on road plane using several parameters such as, steering angle and camera tilt angle. Stand still objects are automatically detected by analyzing the difference between a model flow field and a real optical flow field which is acquired by an in-vehicle camera image. The algorithms have been installed in a desktop PC by Microsoft Visual C++6.0 programs. Experimental results by several kinds of outdoor scenes have shown the effectiveness of the proposed method. Total detection ratio is 91.0% and that in Dangerous case is 100%.

動きベクトルを用いた車載カメラ映像に含まれる雨滴の抽出に関する一考察(高精細度画像の処理・表示および一般)

動きベクトルを用いた車載カメラ映像に含まれる雨滴の抽出に関する一考察(高精細度画像の処理・表示および一般)