Region Of Interest Detection For Automotive Applications

Browne, Lorcan; Jones, Edward; Glavin, Martin

doi:10.4108/icst.isvcs2008.3861

Cited by 1 publication

(4 citation statements)

References 9 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The suggested method is based on video obtained from a single forward-facing, standard optical camera. The approach is to first extract information about objects moving in the field of view of the camera, then to further process this information to determine which regions contain objects that may pose the "Regions of Interest" [3]. The flowchart in Figure 1 shows the different stages of the algorithm.…”

Section: Region Of Interestmentioning

confidence: 99%

“…The components of the algorithm are described further in the following sub-sections. [3]. On this ROI, Image subtraction is carried out to find a moving vehicle.…”

Section: Region Of Interestmentioning

confidence: 99%

“…In scenarios where the image capture cameras have a constant focal length, this strategy is helpful, and Don't tilt or pan, so maintaining the camera's focus on a moving foreground and background. There are a variety of methods for removing the best backdrop from the incoming image stream to retrieve the desired objects so that they can be processed further [3].…”

Section: Background Image Subtractionmentioning

confidence: 99%

“…There are two regions: the regions you are interested in and the region you are not interested in. If you want to mask out the region that is the region you are not interested in, then complement the mask of the ROI[3]. On this ROI, Image subtraction is carried out to find a moving vehicle.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Applying Machine Learning and Object Detection Foridentifying Traffic Rule Violation: Over Speeding

Chauhan,

Rahim,

Kurani

et al. 2023

IJEAST

View full text Add to dashboard Cite

We have observed that traffic offences are manually inspected by the traffic department in many developing nations, such as India. Such systems enable traffic management and traffic law enforcement. Following this, the present study focuses on Traffic violations as one of the key issues hindering the efficient operations and safety aspects in the urban traffic area. Currently, traffic violations are controlled using enforcement measure that demands a high manual workforce, and resource which entails high economic cost. Contrary, the use of automatic traffic violation detection systems in many of the urban cities of the developed and developing countries is being implemented for a higher degree of traffic enforcement and adherence to the traffic operation rules. Following this the present study focus on the applicationof machine learning for automatic traffic violation detection for weak- lane disciplined mixed traffic conditions. The resulting exhibit that the developed model can detect traffic violations of over-speeding at an accuracy of 95%. The traditional methods like speed camera use the basic principle of the Doppler Effect and RADAR technologies but at present,Due to their drawbacks which vary from the expense of purchasing equipment, increase in work and labour costs, and lack of monitoring, they are ineffective. As a result, it is a straightforward, affordable, and practical speed detecting method in Mixed traffic conditions. So here, we prefer automatic speed violation detection. The suggested solution will be highly accurate and cost-efficient.

show abstract

Section: Region Of Interestmentioning

confidence: 99%

“…The components of the algorithm are described further in the following sub-sections. [3]. On this ROI, Image subtraction is carried out to find a moving vehicle.…”

Section: Region Of Interestmentioning

confidence: 99%

Section: Background Image Subtractionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations