Enhancing Light Blob Detection for Intelligent Headlight Control Using Lane Detection

Eum, Sungmin; Jung, Ho Gi

doi:10.1109/tits.2012.2233736

Cited by 46 publications

(28 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many researchers have presented their work in the subject. Some applications use the nonvision based approaches [3], [4] to implement driver assistance systems whereas others use vision based approaches for traffic surveillance and driver assistance systems [5], [6], [7]- [13]. Headlights have been widely used for detecting vehicles at night.…”

Section: Related Work and Implementationsmentioning

confidence: 99%

“… Headlight identification: Various approaches for detecting headlight blob are based on thresholding techniques. The most common among them is fixed thresholding [5][8], [9]or adaptive thresholding [6]. Though thresholding is fast, it is incapable of capturing objects structure.…”

Section: Image Based Approachmentioning

confidence: 99%

“…Though thresholding is fast, it is incapable of capturing objects structure. This leads to use of LoG filters [5]. The result of these methods is extracted light blob.…”

Section: Image Based Approachmentioning

confidence: 99%

“…For nighttime detection, headlights are tracked. The very common approach is using kalman filter [5], [6], [7]. Other forms of tracking are rule based tracking [10] and nearest neighbor matching [11].…”

mentioning

confidence: 99%

See 3 more Smart Citations

Night-time Vehicle Detection for Automatic Headlight Beam Control

Sevekar¹,

Dhonde²

2017

IJCA

View full text Add to dashboard Cite

A clear visibility of the road ahead is significant concern for safe nighttime driving. However, high beams are utilized less than sufficient on the roads since drivers are afraid of dazzling others. Subsequently, the smart programmed control of vehicles front lamp is of incredible significance.According to road accident surveys, majority of the accidents occur in dark. Visibility in dark is significant issue for safe driving. Therefore careless drivers continue using a high beam even though approaching vehicle is observed. These high beams create glare for approaching drivers which then causes temporary blindness. To solve this problem, nighttime vehicle detection has a great importance. This paper reviews various attempts made to solve the problem, need of study, currently implemented relevant systems and related work, different approaches to solve problem and various applications. The survey shows sensor based approaches cannot be used to satisfy real time requirements of a system. However, a simple and reliable solution needs to be developed so that, it can be implemented in each vehicle. An attempt has been made to present an image processing based system that detects vehicles and selects ideal beam so that accidents due to temporary blindness can be reduced. General TermsLight blob detection, Support vector machines.

show abstract

Section: Related Work and Implementationsmentioning

confidence: 99%

Section: Image Based Approachmentioning

confidence: 99%

“…Though thresholding is fast, it is incapable of capturing objects structure. This leads to use of LoG filters [5]. The result of these methods is extracted light blob.…”

Section: Image Based Approachmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Night-time Vehicle Detection for Automatic Headlight Beam Control

Sevekar¹,

Dhonde²

2017

IJCA

View full text Add to dashboard Cite

show abstract

“…To investigate whether the target implementations work well with a multiclass problem and a classifier under development, in other words, when the classifier is not finely tuned and the accuracy is not so high, the three implementations are applied to a light-blob classifier [97] for IHC under development.…”

Section: Application To Intelligent Headlight Controlmentioning

confidence: 99%

Support Vector Number Reduction: Survey and Experimental Evaluations

Jung

Kim

2014

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

Abstract-Although a support vector machine (SVM) is one of the most frequently used classifiers in the field of intelligent transportation systems and shows competitive performances in various problems, it has the disadvantage of requiring relatively large computations in the testing phase. To make up for this weakness, diverse methods have been researched to reduce the number of support vectors determining the computations in the testing phase. This paper is intended to help engineers using the SVM to easily apply support vector number reduction to their own particular problems by providing a state-of-the-art survey and quantitatively comparing three implementations belonging to postpruning, which exploits the result of a standard SVM. In particular, this paper confirms that the support vector number of a pedestrian classifier using a histogram-of-oriented-gradientbased feature and a radial-basis-function-kernel-based SVM can be reduced by more than 99.5% without any accuracy degradation using iterative preimage addition, which can be downloaded from the Internet.Index Terms-Reduced-set method, support vector machine (SVM), support vector number reduction (SVNR).

show abstract

Pre‐training Dataset Generation Using Visual Explanation for Classifying Beam of Vehicle Headlights from Nighttime Camera Image

Oyabu

Sultana

Sakagawa

et al. 2021

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

For Automatic High Beam (AHB), the authors are working on a study to classify beam of car headlights as 'High' and 'Low' by realizing the presence of vehicle on road from nighttime in-vehicle camera image using deep learning. In general, weights of deep learning model are pre-trained on a large dataset which are then used as initial weights for other tasks such as image recognition. However, it has been reported that even if the pre-trained weights are used as the initial weights, it is not a factor of accuracy improvement when domain differs. It is important to pre-train a model on a dataset that is suitable for target domain. Therefore, we propose a method to generate a pre-training dataset that can be easily created by using visual explanation which represents where a deep learning model is looking at when performs a task. Then, we applied the proposed pre-training dataset on the headlight beam classification in the nighttime in-vehicle camera image and verified the effectiveness of our method.

show abstract

Enhancing Light Blob Detection for Intelligent Headlight Control Using Lane Detection

Cited by 46 publications

References 24 publications

Night-time Vehicle Detection for Automatic Headlight Beam Control

Night-time Vehicle Detection for Automatic Headlight Beam Control

Support Vector Number Reduction: Survey and Experimental Evaluations

Pre‐training Dataset Generation Using Visual Explanation for Classifying Beam of Vehicle Headlights from Nighttime Camera Image

Contact Info

Product

Resources

About