Apple Detection in Natural Environment Using Deep Learning Algorithms

Xuan, Guantao; Gao, Chong; Shao, Yuanyuan; Zhang, Meng; Wang, Yongxian; Zhong, Jingrun; Li, Qingguo; Peng, Hongxing

doi:10.1109/access.2020.3040423

Cited by 29 publications

(10 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The authors of [25] show that customized training and the use of image augmentation [26] lead to an increase in the quality of such systems. Xuan G. et al [27] achieve f-measure indices up to 91-94% under different illumination conditions on green apples and 94-95% on red apples. The authors of [28], in addition to apples, add pears to the recognition system based on YOLOv3 and suggest using Kalman filtering for tracking fruits while moving.…”

Section: Related Workmentioning

confidence: 99%

Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415

et al. 2022

View full text Add to dashboard Cite

Despite the great possibilities of modern neural network architectures concerning the problems of object detection and recognition, the output of such models is the local (pixel) coordinates of objects bounding boxes in the image and their predicted classes. However, in several practical tasks, it is necessary to obtain more complete information about the object from the image. In particular, for robotic apple picking, it is necessary to clearly understand where and how much to move the grabber. To determine the real position of the apple relative to the source of image registration, it is proposed to use the Intel Real Sense depth camera and aggregate information from its depth and brightness channels. The apples detection is carried out using the YOLOv3 architecture; then, based on the distance to the object and its localization in the image, the relative distances are calculated for all coordinates. In this case, to determine the coordinates of apples, a transition to a symmetric coordinate system takes place by means of simple linear transformations. Estimating the position in a symmetric coordinate system allows estimating not only the magnitude of the shift but also the location of the object relative to the camera. The proposed approach makes it possible to obtain position estimates with high accuracy. The approximate root mean square error is 7–12 mm, depending on the range and axis. As for precision and recall metrics, the first is 100% and the second is 90%.

show abstract

Section: Related Workmentioning

confidence: 99%

Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415

et al. 2022

View full text Add to dashboard Cite

show abstract

“…As shown in Equation 5, the location loss function reflects the correlation between predicted boxes and labeled boxes concerning the match with one category, adopting the same loss function as Faster R-CNN [21].…”

Section: Cm-ghostnet-ssd Loss Functionmentioning

confidence: 99%

Research on Pet Recognition Algorithm With Dual Attention GhostNet-SSD and Edge Devices

Hou

Meng

2022

IEEE Access

View full text Add to dashboard Cite

The ability to identify pets via smart home technology is crucial for managing pet health. However, the present popular object detection method model is too large, support for edge devices is insufficient, and performance is subpar for small object identification in the traditional algorithm, such as SSD. In this study, we propose an improved SSD object detection network algorithm based on a GhostNet network with a dual attention mechanism. A ghost module is introduced to create a lightweight ghost network based on the SSD network. The ghost module is combined with the ECA attention mechanism to dynamically allocate parameters to the target and alter the detection region weights to enhance the model performance. In order to reinforce the model and increase the precision of the small target, the CBAM module was also introduced. It assigns major weights to the output target regions of the large-scale feature layer. According to the experimental findings, the CBAM-GhostNet-SSD network significantly reduces the number of backbone parameters and the calculation amount is reduced by 98.23% when compared to the classic SSD. This lays the groundwork for the self-developed algorithm's successful deployment of edge devices. The predicted rate synchronization increased by 3.1 times, also mAP was 14.5% higher than SSD. The lightweight model quantitative transformation utilized in edge equipment can accurately evaluate the target region and realize dynamic detection. This has a certain guiding significance for the subsequent detection of household pet targets.INDEX TERMS deep learning, object detection, ghostnet, attention modules, edge devices.

show abstract

“…The content of the pixel will be determined by comparing these values to the database [21], [22]. From the figure 2.3 and 2.4 the input image sample chosen as 5 * 5 dimension with filter 3 * 3 dimension.…”

Section: Cnn -General Working Principlementioning

confidence: 99%

A New Improved Binary Convolutional Model for Classification of Images

Hemalatha

Shankar²,

Raj

2022

SCPE

View full text Add to dashboard Cite

There are numerous image classification strategies are developed in deep learning. However, due to the complexity of images, conventional image classification strategies have been incapable to meet real application needs. As the amount of pixel information rises, the classification becomes more difficult. However, CNN is widely used method for object identification in picture due to its simple and accurate, but still, it remains hazy which strategies are most supportive for analysing and distinguishing the objects in pictures. In this paper we introduced a CNN network and clustering-based technique called IBCNN to perform classification based on patch extraction. The proposed method can accomplish their goals in the following four different ways: a) Automatic Kernel selection; b) resilient patch size selection; c) CNN layer; and d) pooling layer modification. In addition, it also modifies the pooling layer with average value and calculate the pixel size. The proposed method was applied on ten different image datasets. Finally, the proposed model is compared to three benchmarking models: such as WCNN, MLP, and ELM-CNN to estimate its performance. The obtained results shows that the proposed method gives competitive results compared to the other models.

show abstract

Apple Detection in Natural Environment Using Deep Learning Algorithms

Cited by 29 publications

References 27 publications

Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415

Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415

Research on Pet Recognition Algorithm With Dual Attention GhostNet-SSD and Edge Devices

A New Improved Binary Convolutional Model for Classification of Images

Contact Info

Product

Resources

About