A review of research on object detection based on deep learning

Deng, Jun; Xuan, Xiaojing; Wang, Weifeng; Zhao, Li; Yao, Hanwen; Wang, Zhiqiang

doi:10.1088/1742-6596/1684/1/012028

Cited by 67 publications

(42 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Unlike the two-stage detection models, the YOLOv1 detection model has a simple CNN network structure without the extraction process of region proposal. It uses the entire graph as input of the network that outputs the location and category of the bounding box [ 37 ]. YOLOv2 uses Darknet-19 for fully convolutional feature extraction and anchor box mechanism, k-means clustering and multi-scale training, which improves recall and accuracy.…”

Section: Literature Reviewmentioning

confidence: 99%

“…SSD combines the concept of regression in the YOLO algorithm and anchor box in Faster R-CNN [ 35 ]. YOLOv4 uses CSPDarknet53 as the backbone network and adds weighted residual connection, cross stage partial connection, cross mini batch normalization, self-adversarial training, mish activation, mosaic data augmentation, dropblock and complete intersection over union (CIoU) to the original YOLO framework [ 37 ]. In the YOLO series, YOLOv5 is the latest and improved version in terms of speed, size and accuracy.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the Forest Ecosystem through a Semi-Autonomous Quadruped Robot and a Hexacopter UAV

Idrissi

Hussain

Barua

et al. 2022

Sensors

View full text Add to dashboard Cite

Accurate and timely monitoring is imperative to the resilience of forests for economic growth and climate regulation. In the UK, forest management depends on citizen science to perform tedious and time-consuming data collection tasks. In this study, an unmanned aerial vehicle (UAV) equipped with a light sensor and positioning capabilities is deployed to perform aerial surveying and to observe a series of forest health indicators (FHIs) which are inaccessible from the ground. However, many FHIs such as burrows and deadwood can only be observed from under the tree canopy. Hence, we take the initiative of employing a quadruped robot with an integrated camera as well as an external sensing platform (ESP) equipped with light and infrared cameras, computing, communication and power modules to observe these FHIs from the ground. The forest-monitoring time can be extended by reducing computation and conserving energy. Therefore, we analysed different versions of the YOLO object-detection algorithm in terms of accuracy, deployment and usability by the EXP to accomplish an extensive low-latency detection. In addition, we constructed a series of new datasets to train the YOLOv5x and YOLOv5s for recognising FHIs. Our results reveal that YOLOv5s is lightweight and easy to train for FHI detection while performing close to real-time, cost-effective and autonomous forest monitoring.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Evaluating the Forest Ecosystem through a Semi-Autonomous Quadruped Robot and a Hexacopter UAV

Idrissi

Hussain

Barua

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The target detection algorithm [ 39 ] is suitable to solve these problems. Regression‐based target detection algorithms are called one‐stage models because they omit the candidate region generation step and directly implement feature extraction, target classification, and target regression in the same convolutional neural network.…”

Section: Literature Reviewmentioning

confidence: 99%

Real‐Time and Image‐Based AQI Estimation Based on Deep Learning

Zhang

Tian

et al. 2022

Advcd Theory and Sims

View full text Add to dashboard Cite

The real-time information on surrounding air quality index (AQI) is important for the public to protect themselves from air pollution. Traditional methods have some shortages regarding the estimation time and running efficiency. Consequently, the AQI results cannot meet the needs of personal protection and environmental management. With the popularity of smart terminals, it is easier to collect particular environmental images for AQI estimation tasks. Therefore, a real-time and image-based deep learning model named YOLO-AQI is proposed. Based on the object detection algorithms, the model has better performance regarding the AQI estimation speed. By optimizing the parameter transfer and network structure, the model takes an average of 0.0582 s to perform feature analysis and achieves 75.15% accuracy on AQI estimation tasks. Comparing YOLO-AQI with several image recognition models (VGG, AlexNet, GoogLeNet, MobileNet, and ResNet), it shows that YOLO-AQI outperforms other models by 14.8% on accuracy and by 71.1% on running speed. This method can provide real-time AQI level information for remote areas such as rural.

show abstract

“…Object detection is a fundamental study area in computer vision, artificial intelligence, and other related domains. The main goal of object detection is to find the target of interest in an image, establish the category, and provide the bounding box for each target (Deng et al 2020). In our study, we have primarily transformed our predicted output image into a grayscale image and then thresholded it to identify the region of interest.…”

Section: Image Processing Using Computer Vision Approachmentioning

confidence: 99%

Detection of Epithelial Giant Cells in Nasal Aspirate Cytological Smears Using Deep Learning and Computer Vision Techniques: an Approach for Early Diagnosis of Measles Disease

Rao¹,

Pai²,

Nayak³

et al. 2022

Philipp J Sci

View full text Add to dashboard Cite

Measles disease is usually diagnosed through an RT-PCR test, which detects the measles-specific IgM antibody in serum samples. However, this approach is highly expensive and requires a lot of sample preprocessing making it a laborious task. Another approach for measles identification could be a cytological smear test – which is a low-cost, quick, and accurate method for the early detection of measles disease. Due to the lack of experience in interpreting smears, its usage in clinical practice is currently limited. We have designed a MesoSpot, an automatic segmentation method for the first time that can precisely define the region of giant cells in nasal aspirate cytological smears and assist medical experts in the speedy diagnosis of measles disease. MesoSpot was developed by training 500 cytological smear images collected from various web sources. We have used 50 cytological smear images to validate the model and evaluated its performance using various metrics. The main purpose of the proposed study is to detect multinucleated giant cells and estimate their size and number, which aids in assessing the disease's severity. To annotate the giant cells in each cytological smear, we have applied a semantic segmentation approach using APEER software. UNET architecture was used to train the model, and image processing techniques were used to spot giant cells and predict their size. Various performance metrics were employed to assess the model's efficiency. The model's output was also compared to the ground truth image and statistically confirmed.

show abstract

A review of research on object detection based on deep learning

Cited by 67 publications

References 5 publications

Evaluating the Forest Ecosystem through a Semi-Autonomous Quadruped Robot and a Hexacopter UAV

Evaluating the Forest Ecosystem through a Semi-Autonomous Quadruped Robot and a Hexacopter UAV

Real‐Time and Image‐Based AQI Estimation Based on Deep Learning

Detection of Epithelial Giant Cells in Nasal Aspirate Cytological Smears Using Deep Learning and Computer Vision Techniques: an Approach for Early Diagnosis of Measles Disease

Contact Info

Product

Resources

About