People Detection and Finding Attractive Areas by the use of Movement Detection Analysis and Deep Learning Approach

Kajabad, Ebrahim Najafi; Ivanov, Sergey V.

doi:10.1016/j.procs.2019.08.209

Cited by 35 publications

(24 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, even the latest version of YOLOv3 has some limitations. If there are two anchor boxes but three objects in the same grid cell, it does not support them correctly, which ultimately leads to missing objects (Kajabad and Ivanov, 2019). YOLO achieves about 10% missing detection rate for pedestrian detection (Lan et al, 2018).…”

Section: Deep Learningmentioning

confidence: 99%

“…Proper detection of people is crucial for autonomous cars, advertising planning and many other industries and public safety. Kajabad and Ivanov (2019) proposed a method of finding areas more attractive to customers (hot zones) based on people detection. Sometimes, people must be detected in a heavy industry environment (Zengeler et al, 2019) or in hazy weather (Li et al, 2019).…”

Section: People Detectionmentioning

confidence: 99%

“…These two-step detection algorithms divide the problem into two stages: (i) generating region proposals and (ii) classification of candidate regions. But these traditional deep learning algorithms suffer from low speed (Kajabad and Ivanov, 2019). To overcome this limitation, Redmon et al (2016) proposed a one-step detection algorithm called YOLO (You Only Look Once), enabling easy implementation end-to-end object detection.…”

Section: Deep Learningmentioning

confidence: 99%

“…Darknet architecture is a pre-trained model for classifying 80 different classes. Several improvements to the YOLO model have been proposed for detecting people (Putra et al, 2017(Putra et al, , 2018Lan et al, 2018;He et al, 2019;Li et al, 2019), but even standard YOLOv3 outperforms traditional computer vision methods and most of deep neural network methods (Ghosh and Das, 2019;Zengeler et al, 2019;Kajabad and Ivanov, 2019;Yun et al, 2018).…”

Section: Deep Learningmentioning

confidence: 99%

“…There are mainly three types of methods to count the number of people in the crowd from video (Ren et al, 2018): (i) statistical method to estimate the number of people in a region, (ii) combination of object detection with object tracking and (iii) use of path information of the points, with subsequent cluster analysis of the feature point path. People detection in crowded spaces is the most challenging task, because of the people occlusions (Stewart et al, 2016;Kajabad and Ivanov, 2019). Crowd counting requires development of new methods (Stewart et al, 2016;Lei et al, 2020) like Dynamic Region Division (He et al, 2019).…”

Section: Crowd Countingmentioning

confidence: 99%

See 4 more Smart Citations

Peer Review #1 of "Analysis of pedestrian activity before and during COVID-19 lockdown, using webcam time-lapse from Cracow and machine learning (v0.1)"

2020

View full text Add to dashboard Cite

, COVID-19 pandemic reached Europe. Many countries, including Poland imposed lockdown as a method of securing social distance between potentially infected. Stay-at-home orders and movement control within public space not only affected the touristm industry, but also the everyday life of the inhabitants. The hourly time-lapse from four HD webcams in Cracow (Poland) are used in this study to estimate how pedestrian activity changed during COVID-19 lockdown. The collected data covers the period from June 9, 2016 to April 19, 2020 and comes from various urban zones. One zone is tourist, one is residential and two are mixed. In the first stage of the analysis, state-of-the-art machine learning algorithm (YOLOv3) is used to detect people. Additionally, a non-standard application of the YOLO method is proposed, oriented to the images from HD webcams. This approach (YOLOtiled) is less prone to pedestrian detection errors with the only drawback being the longer computation time. Splitting the HD image into smaller tiles increases the number of detected pedestrians by over 50%. In the second stage, the analysis of pedestrian activity before and during the COVID-19 lockdown is conducted for hourly, daily and weekly averages. Depending on the type of urban zone, the number of pedestrians decreased from 33% in residential zones to 85% in tourist zones located in the Old Town. The presented method allows for more efficient detection and counting of pedestrians from HD time-lapse webcam images compared to SSD, YOLOv3 and Faster R-CNN. The result of the research is a published database with the detected number of pedestrians from the four-year observation period for four locations in Cracow.

show abstract

Section: Deep Learningmentioning

confidence: 99%