scholarly journals End-to-End Monocular Range Estimation for Forward Collision Warning

Sensors ◽  
2020 ◽  
Vol 20 (20) ◽  
pp. 5941
Author(s):  
Jie Tang ◽  
Jian Li

Estimating range to the closest object in front is the core component of the forward collision warning (FCW) system. Previous monocular range estimation methods mostly involve two sequential steps of object detection and range estimation. As a result, they are only effective for objects from specific categories relying on expensive object-level annotation for training, but not for unseen categories. In this paper, we present an end-to-end deep learning architecture to solve the above problems. Specifically, we represent the target range as a weighted sum of a set of potential distances. These potential distances are generated by inverse perspective projection based on intrinsic and extrinsic camera parameters, while a deep neural network predicts the corresponding weights of these distances. The whole architecture is optimized towards the range estimation task directly in an end-to-end manner with only the target range as supervision. As object category is not restricted in the training stage, the proposed method can generalize to objects with unseen categories. Furthermore, camera parameters are explicitly considered in the proposed method, making it able to generalize to images taken with different cameras and novel views. Additionally, the proposed method is not a pure black box, but provides partial interpretability by visualizing the produced weights to see which part of the image dominates the final result. We conduct experiments to verify the above properties of the proposed method on synthetic and real-world collected data.

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Ki-Yeong Park ◽  
Sun-Young Hwang

We propose a range estimation method for vision-based forward collision warning systems with a monocular camera. To solve the problem of variation of camera pitch angle due to vehicle motion and road inclination, the proposed method estimates virtual horizon from size and position of vehicles in captured image at run-time. The proposed method provides robust results even when road inclination varies continuously on hilly roads or lane markings are not seen on crowded roads. For experiments, a vision-based forward collision warning system has been implemented and the proposed method is evaluated with video clips recorded in highway and urban traffic environments. Virtual horizons estimated by the proposed method are compared with horizons manually identified, and estimated ranges are compared with measured ranges. Experimental results confirm that the proposed method provides robust results both in highway and in urban traffic environments.


2016 ◽  
Vol 17 (4) ◽  
pp. 1157-1167 ◽  
Author(s):  
Jianqiang Wang ◽  
Chenfei Yu ◽  
Shengbo Eben Li ◽  
Likun Wang

2021 ◽  
Author(s):  
yang teng ◽  
Shupei TANG ◽  
lai heda meng ◽  
Liji Wu ◽  
Zhiqing HAN ◽  
...  

Abstract Home range size estimation is a crucial basis for developing effective conservation strategies and provides important insights into animal behavior and ecology. This study aimed at analyzing the home range variations, the influence of altitude in habitat selection, and comparing three methods in home range estimation of Chinese gorals (Naemorhedus griseus) living at a cliff landscape. The results indicated that there were significant differences between the annual home range sizes of individual animals but there was no difference in their seasonal home range sizes based on GPS tracking data of five female Chinese gorals from February 2015 to September 2018. The monthly home ranges decreased dramatically in May, June and July due to birth-giving. Notable seasonal variations were found in the micro-habitats of the Chinese gorals, as reflected by the altitude they inhabit, with higher altitude habitats used in spring and lower altitude habitats used in winter. Additionally, the altitude of monthly habitats was lowest in January, which may indicate an adaptation to low air temperature. We also found differences between estimation methods, namely minimum convex polygon (MCP), kernel density estimation (KDE) and α-local convex hull (α-LoCoH), with seasonal home range sizes derived from α-LoCoH being substantially smaller than those derived from MCP and KDE. In conclusion, our findings filled the gaps in home range study for this endangered species and contributed to effective conservation strategies. Considerations shall have to be given to the variations in home range estimation caused by different methods when dealing with rugged habitats, so as to make sure that any interpretation concerning the habitat use of the targeted species made on basis of such results would be meaningful and valid.


2021 ◽  
Author(s):  
Yang Teng ◽  
Shupei TANG ◽  
Dalai Menghe ◽  
Liji Wu ◽  
Zhiqing HAN ◽  
...  

Abstract Home range size estimation is a crucial basis for developing effective conservation strategies and provides important insights into animal behavior and ecology. This study aimed at analyzing the home range variations, the influence of altitude in habitat selection, and comparing three methods in home range estimation of Chinese gorals (Naemorhedus griseus) living at a cliff landscape. The results indicated that there were significant differences between the annual home range sizes of individual animals but there was no difference in their seasonal home range sizes based on GPS tracking data of five female Chinese gorals from February 2015 to September 2018. The monthly home ranges decreased dramatically in May, June and July due to birth-giving. Notable seasonal variations were found in the micro-habitats of the Chinese gorals, as reflected by the altitude they inhabit, with higher altitude habitats used in spring and lower altitude habitats used in winter. Additionally, the altitude of monthly habitats was lowest in January, which may indicate an adaptation to low air temperature. We also found differences between estimation methods, namely minimum convex polygon (MCP), kernel density estimation (KDE) and α-local convex hull (α-LoCoH), with seasonal home range sizes derived from α-LoCoH being substantially smaller than those derived from MCP and KDE. In conclusion, our findings filled the gaps in home range study for this endangered species and contributed to effective conservation strategies. Considerations shall have to be given to the variations in home range estimation caused by different methods when dealing with rugged habitats, so as to make sure that any interpretation concerning the habitat use of the targeted species made on basis of such results would be meaningful and valid.


Sign in / Sign up

Export Citation Format

Share Document