A Review of the Impacts of Defogging on Deep Learning-Based Object Detectors in Self-Driving Cars

Ogunrinde, Isaac Oluwadunsin; Bernadin, Shonda

doi:10.1109/southeastcon45413.2021.9401941

Cited by 14 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…YOLO and its derivatives can instantly predict bounding boxes and the object class after extracting features from an input image. One-stage object detectors generate candidate regions, instantly used to classify and predict the target's spatial location [1]. Backbone networks such as feature pyramid networks (FPNs) [40] together with one-stage detectors such as YOLO [37], YOLO9000 [38], YOLOv3 [39], or SSD [41] were used to detect objects via numerous detection branches in one operation instead of predicting the potential locations and classifying them later.…”

Section: Object Detection By Camera Onlymentioning

confidence: 99%

“…An image model suggested by Koschmieder [62] has been frequently used in the scientific literature [1]:…”

Section: Fog Imaging Modelmentioning

confidence: 99%

“…AVs encounter several difficulties under adverse weather conditions, such as snow, fog, haze, shadow, and rain [1][2][3][4][5][6][7][8]. AVs may suffer from poor decision-making and control if their perception systems are degraded by adverse weather.…”

Section: Introductionmentioning

confidence: 99%

“…These radar and camera problems usually led to inaccuracies in the real time detection of the bounding box of an object or location in an image, especially when the object is not nearby or when the object is too small under medium and heavy foggy weather conditions. Thus, applying single-sensor modal CNN-based object detection algorithms on such distorted data has proven inefficient [1,2].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Deep Camera-Radar Fusion with Attention Framework for Autonomous Vehicle Vision in Foggy Weather Conditions

Ogunrinde¹,

Bernadin²

2023

Preprint

View full text Add to dashboard Cite

AVs suffer reduced maneuverability and performance due to the degradation in sensor performances in fog. Such degradation causes significant object detection errors essential for AVs' safety-critical conditions. For instance, YOLOv5 performs significantly well under favorable weather but suffers miss detections and false positives due to atmospheric scattering caused by fog particles. Existing deep object detection techniques often exhibit a high degree of accuracy. The drawback is being sluggish at object detection in fog. Object detection methods with fast detection speed have been obtained using deep learning at the expense of accuracy. The problem of the lack of balance between detection speed and accuracy in fog persist. This paper presents an improved YOLOv5-based multi-sensor fusion network that combines radar's object detection with a camera image bounding box. We transformed radar detection by mapping the radar detections into a two-dimensional image coordinate and projected the resultant radar image on the camera image. Using the attention mechanism, we emphasized and improved important feature representation used for object detection while reducing high-level feature information loss. We trained and tested our multi-sensor fusion network on clear and multi-fog weather datasets obtained from the CARLA simulator. Our result shows that the proposed method significantly enhances the detection of distant and small objects. Our small CR-YOLOnet model best strikes a balance between accuracy and speed with an accuracy of 0.849 at 69 fps.

show abstract

Section: Object Detection By Camera Onlymentioning

confidence: 99%

“…An image model suggested by Koschmieder [62] has been frequently used in the scientific literature [1]:…”

Section: Fog Imaging Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Camera-Radar Fusion with Attention Framework for Autonomous Vehicle Vision in Foggy Weather Conditions

Ogunrinde¹,

Bernadin²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Khan and Ahmed ( 2021 ) developed a novel convolutional neural network structure for detecting vehicle road images limited by weather factors, and the detection speed was significantly improved. Ogunrinde and Bernadin ( 2021 ) used CycleGAN combined with YOLOv3 for the KITTI dataset to improve the detection efficiency of moderate haze images. Wang et al ( 2022 ) proposed a vehicle detection method based on pseudo-visual search and the histogram of oriented gradients (HOG)-local binary pattern feature fusion, which achieved an accuracy of 92.7% and a detection speed of 31 fps.…”

Section: Related Workmentioning

confidence: 99%

Real-time vehicle target detection in inclement weather conditions based on YOLOv4

Wang

Zhao

et al. 2023

Front. Neurorobot.

View full text Add to dashboard Cite

As a crucial component of the autonomous driving task, the vehicle target detection algorithm directly impacts driving safety, particularly in inclement weather situations, where the detection precision and speed are significantly decreased. This paper investigated the You Only Look Once (YOLO) algorithm and proposed an enhanced YOLOv4 for real-time target detection in inclement weather conditions. The algorithm uses the Anchor-free approach to tackle the problem of YOLO preset anchor frame and poor fit. It better adapts to the detected target size, making it suitable for multi-scale target identification. The improved FPN network transmits feature maps to unanchored frames to expand the model's sensory field and maximize the utilization of model feature data. Decoupled head detecting head to increase the precision of target category and location prediction. The experimental dataset BDD-IW was created by extracting specific labeled photos from the BDD100K dataset and fogging some of them to test the proposed method's practical implications in terms of detection precision and speed in Inclement weather conditions. The proposed method is compared to advanced target detection algorithms in this dataset. Experimental results indicated that the proposed method achieved a mean average precision of 60.3%, which is 5.8 percentage points higher than the original YOLOv4; the inference speed of the algorithm is enhanced by 4.5 fps compared to the original, reaching a real-time detection speed of 69.44 fps. The robustness test results indicated that the proposed model has considerably improved the capacity to recognize targets in inclement weather conditions and has achieved high precision in real-time detection.

show abstract

Enhancing the image quality of blue light cystoscopy through green-hue correction and fogginess removal

Chang,

Bermoy,

Chang

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Blue light cystoscopy (BLC) is a guideline-recommended endoscopic tool to detect bladder cancer with high sensitivity. Having clear, high-quality images during cystoscopy is crucial to the sensitive, efficient detection of bladder tumors; yet, important diagnostic information is often missed or poorly visualized in images containing illumination artifacts or impacted by impurities in the bladder. In this study, we introduce computational methods to remove two common artifacts in images from BLC videos: green hue and fogginess. We also evaluate the effect of artifact removal on the perceptual quality of the BLC images through a survey study and computation of Blind/Referenceless Image Spatial Quality Evaluator scores on the original and enhanced images. We show that corrections and enhancements made to cystoscopy images resulted in a better viewing experience for clinicians during BLC imaging and reliably restored lost tissue features that were important for diagnostics. Incorporating these enhancements during clinical and OR procedures may lead to more comprehensive tumor detection, fewer missed tumors during TURBT procedures, more complete tumor resection and shorter procedure time. When used in off-line review of cystoscopy videos, it may also better guide surgical planning and allow more accurate assessment and diagnosis.

show abstract

A Review of the Impacts of Defogging on Deep Learning-Based Object Detectors in Self-Driving Cars

Cited by 14 publications

References 18 publications

Deep Camera-Radar Fusion with Attention Framework for Autonomous Vehicle Vision in Foggy Weather Conditions

Deep Camera-Radar Fusion with Attention Framework for Autonomous Vehicle Vision in Foggy Weather Conditions

Real-time vehicle target detection in inclement weather conditions based on YOLOv4

Enhancing the image quality of blue light cystoscopy through green-hue correction and fogginess removal

Contact Info

Product

Resources

About