Overlapped Apple Fruit Yield Estimation using Pixel Classification and Hough Transform

Kanwal, Zartash; Basit, Abdul; Jawad, Muhammad; Ullah, Ihsan; Sanjrani, Anwar Ali

doi:10.14569/ijacsa.2019.0100271

Cited by 3 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, in fruit identification, if the proposed method is applied to images with denser fruit, it becomes difficult to identify the same fruit because the distance filter may not work as intended. Furthermore, previous studies on apple detection using the Hough transform [24,25] reported that the precisions were 93.5% and 92.0%, respectively, and the proposed method shows a higher value than those reported. The proposed method is effective when applied to data with large volumes and diverse state changes, such as time-series images, because the Hough transform does not detect fruit correctly with fixed parameters.…”

Section: Verifying the Accuracy Of Fruit Detectionmentioning

confidence: 63%

Real-Time Prediction of Growth Characteristics for Individual Fruits Using Deep Learning

Hondo

Kobayashi

Aoyagi

2022

Sensors

View full text Add to dashboard Cite

Understanding the growth status of fruits can enable precise growth management and improve the product quality. Previous studies have rarely used deep learning to observe changes over time, and manual annotation is required to detect hidden regions of fruit. Thus, additional research is required for automatic annotation and tracking fruit changes over time. We propose a system to record the growth characteristics of individual apples in real time using Mask R-CNN. To accurately detect fruit regions hidden behind leaves and other fruits, we developed a region detection model by automatically generating 3000 composite orchard images using cropped images of leaves and fruits. The effectiveness of the proposed method was verified on a total of 1417 orchard images obtained from the monitoring system, tracking the size of fruits in the images. The mean absolute percentage error between the true value manually annotated from the images and detection value provided by the proposed method was less than 0.079, suggesting that the proposed method could extract fruit sizes in real time with high accuracy. Moreover, each prediction could capture a relative growth curve that closely matched the actual curve after approximately 150 elapsed days, even if a target fruit was partially hidden.

show abstract

Section: Verifying the Accuracy Of Fruit Detectionmentioning

confidence: 63%

Real-Time Prediction of Growth Characteristics for Individual Fruits Using Deep Learning

Hondo

Kobayashi

Aoyagi

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…For grape picking, Rodrigo et al selected HOG (Histogram of Oriented Gradients) and LBP (Local Binary Pattern) to extract shape and texture features of grapes, and then used SVM-RBF to build a grape recognition classifier (Perez-Zavala et al, 2018). For apple harvesting, Zartash et al used HS model to locate and segment the apple images, and then used refinement denoising and Hough transform to realize accurate location of the apples (Kanwal et al, 2019). Gu Suhang et al introduced the ASIFT feature to repair the target hollow areas generated by K-means clustering, and used the gPb contour detector and the dynamic threshold Otsu method successively to generate clear and continuous target contours (Gu et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

EBE-YOLOv4: A lightweight detecting model for pine cones in forest

Zhang

Jiang

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

Pine cones are important forest products, and the picking process is complex. Aiming at the multi-objective and dispersed characteristics of pine cones in the forest, a machine vision detection model (EBE-YOLOV4) is designed to solve the problems of many parameters and poor computing ability of the general YOLOv4, so as to realize rapid and accurate recognition of pine cones in the forest. Taking YOLOv4 as the basic framework, this method can realize a lightweight and accurate recognition model for pine cones in forest through optimized design of the backbone and the neck networks. EfficientNet-b0 (E) is chosen as the backbone network for feature extraction to reduce parameters and improve the running speed of the model. Channel transformation BiFPN structure (B), which improves the detection rate and ensures the detection accuracy of the model, is introduced to the neck network for feature fusion. The neck network also adds a lightweight channel attention ECA-Net (E) to solve the problem of accuracy decline caused by lightweight improvement. Meanwhile, the H-Swish activation function is used to optimize the model performance to further improve the model accuracy at a small computational cost. 768 images of pine cones in forest were used as experimental data, and 1536 images were obtained after data expansion, which were divided into training set and test set at the ratio of 8:2. The CPU used in the experiment was Inter Core i9-10885@2.40Ghz, and the GPU was NVIDIA Quadro RTX 5000. The performance of YOLOv4 lightweight design was observed based on the indicators of precision (P), recall (R) and detection frames per second (FPS). The results showed that the measurement accuracy (P) of the EBE-YOLOv4 was 96.25%, the recall rate (F) was 82.72% and the detection speed (FPS) was 64.09F/S. Compared with the original YOLOv4, the precision of detection had no significant change, but the speed increased by 70%, which demonstrated the effectiveness of YOLOv4 lightweight design.

show abstract

“…Various research studies conducted on this area to make video surveillance [3] more versatile and reliable but the detection of suspicious objects is still a challenging job in video surveillance. We need a system that distinguishes and identify highly hazardous situations and makes alerts to take proper action.…”

Section: Introductionmentioning

confidence: 99%

Abandoned Object Detection using Frame Differencing and Background Subtraction

Din¹,

Bashir²,

Basit³

et al. 2020

IJACSA

Self Cite

View full text Add to dashboard Cite

Tracking objects over fixed surveillance cameras are widely used for security purposes in public areas such as train stations, airports, parking areas, and public transportation for the prevention of terrorism. Once the object is accurately detected in the image scene, we can use various visual algorithms to find a number of applications. In this paper, we introduce a model for tracking the multiple objects along with detecting the abandoned luggage in the real time environment. In our model, we used the initial frames to model the background scene. Next, we used the motion model that is background subtraction to detect and track moving objects such as the owner and the luggage. The proposed model also maintains the position history of moving objects followed by the frame differencing technique to find out the luggage history and detect the abandoned luggage by a human. We have used PETS2006 and PETS2007 dataset for the testing of the proposed system in various indoor and outdoor environments with varying lighting conditions.

show abstract

Overlapped Apple Fruit Yield Estimation using Pixel Classification and Hough Transform

Cited by 3 publications

References 11 publications

Real-Time Prediction of Growth Characteristics for Individual Fruits Using Deep Learning

Real-Time Prediction of Growth Characteristics for Individual Fruits Using Deep Learning

EBE-YOLOv4: A lightweight detecting model for pine cones in forest

Abandoned Object Detection using Frame Differencing and Background Subtraction

Contact Info

Product

Resources

About