Implemetation of image classification CNN using multi thread GPU

Han, Seong-Hyeon; Lee, Kwang-Yeob

doi:10.1109/isocc.2017.8368904

Cited by 13 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When acquiring video images, we will encounter a challenging problem: Due to the slow network speed, it takes more time to load the camera video stream, which will put pressure on the time required for subsequent models to identify cyanobacteria blooms, and it may be impossible to fully obtain the cyanobacteria blooms distribution information in the Chaohu Lake lakeshore zone within one hour. To solve this problem, the multithreading technology can set multiple thread plans to enable tasks to process processes in an asynchronous manner [33,34]. The management of the number of threads can rely on the thread pool.…”

Section: Multi-thread Mechanism Constructionmentioning

confidence: 99%

Real-Time Identification of Cyanobacteria Blooms in Lakeshore Zone Using Camera and Semantic Segmentation: A Case Study of Lake Chaohu (Eastern China)

Wang

Qiu

et al. 2023

Sustainability

View full text Add to dashboard Cite

The surface water in the lakeshore zone is the primary area where cyanobacteria bloom floats intensively. In lake water environment monitoring, it has become pressing to accurately identify the distribution and accumulation coverage area of cyanobacteria blooms in the surface water of the lakeshore zone. This study proposes a real-time and dynamic monitoring technology for cyanobacteria blooms in surface water using a shore-based camera monitoring network. The specific work is as follows: Chaohu Lake, a large eutrophic lake in China, is selected as the research object. The multithreading technology is used to dynamically obtain the hourly video images of 43 cameras around Chaohu Lake. The semantic segmentation method is used to identify the cyanobacteria blooms in the video images, calculate the coverage of cyanobacteria blooms, and draw the spatial distribution map of cyanobacteria blooms in the lakeshore zone of Chaohu Lake. To improve the accuracy of cyanobacteria blooms recognition, we use the ResNet-50 network to integrate three semantic segmentation models, namely FCN, U-net, and DeeplabV3+. By comparing the cyanobacteria blooms results identified by the three methods, it is found that the boundary of the cyanobacteria blooms results identified by DeeplabV3+(ResNet-50) is clear, which is more consistent with the real spatial information of the distribution of cyanobacteria blooms and is more suitable for monitoring the hourly dynamic changes of cyanobacteria blooms in the Chaohu Lake lakeshore zone. The results demonstrated that the time requirement of monitoring cyanobacteria blooms in real time on an hourly basis could be met by utilizing technology that uses multiple threads. The OA (Overall Accuracy), MPA (Mean Pixel Accuracy), IOU (Intersection Over Union) of cyanobacteria blooms, and the IOU of water values of the DeeplabV3+(ResNet-50) were the highest, which were 0.83, 0.82, 0.71, and 0.74, and the RMSE between the predicted and real cyanobacterial blooms coverage of 43 cameras was 6.65%. The above values show that DeeplabV3+(ResNet-50) is this technology’s most suitable semantic segmentation model. This technique can provide technical support for the scientific development of a cyanobacteria blooms management plan in the lakeshore zone of Chaohu Lake by calculating the coverage area of cyanobacteria blooms and drawing the spatial distribution map of cyanobacteria blooms in the lakeshore zone.

show abstract

Section: Multi-thread Mechanism Constructionmentioning

confidence: 99%

Real-Time Identification of Cyanobacteria Blooms in Lakeshore Zone Using Camera and Semantic Segmentation: A Case Study of Lake Chaohu (Eastern China)

Wang

Qiu

et al. 2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…In [6] training is performed with a total of 240 images and testing is done on 60 images, with an overall accuracy of 91.6%. Similarly, in [7], it shows the CNN implementation for the CIFAR-10 Image classification performed in multithread GPU. The CIFAR-10 has RGB images similar to images we used.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The CIFAR-10 has RGB images similar to images we used. The architecture used in [7] has used 3 convolutional layers, where the depth of layers is 16,16 and 32 respectively. The final images were applied to a fully connected layer and finally categorized into 10 different classes.…”

Section: Literature Reviewmentioning

confidence: 99%

Automated Agronomy: Evaluation of Fruits Ripeness Using Machine Learning Approach

Chmaj

Sharma

Selvaraj

2021

Proceedings of the 27th International Conference on Systems Engineering, ICSEng 2020

View full text Add to dashboard Cite

Fruit orchards require a lot of tasks, monitoring the current state of fruits and anticipated time of harvest. Precise estimation of harvest time is a key knowledge for the supply chain and timely delivery to grocery stores. The current process relies on human agronomists, who visit orchards often and do the visual evaluation of number of fruits, their ripeness and expected time of harvest. In this paper, we propose a preliminary work for automation of the fruits' evaluation process, using machine learning algorithms to evaluate pictures of the trees. For our current system we have used picturesone picture per tree, to constitute the base for evaluation of each tree using multiple pictures and video captured by drones. In this paper, Convolutional Neural Network (CNN) is used for fruit image classification based on ripeness stage of the fruits. Fruits classification was based on appropriate surface color and shape features and CNN was used to extract these features for classification. The model showed 96.43% accuracy. The ultimate goal of this work is to fully automate the process of orchard trees evaluation, including estimation of number of fruits on the tree (including non-visible ones), their ripeness and time of harvest to address the commercial market delivery planning requirements.

show abstract

“…To determine the hue or tone group mixture that offered the most notable categorization correctness in evaluating the real-ness of wheat flour, the textural properties of individual bits were extracted from various colors and color band combinations of images. hue descriptors interpretation was utilized too to survey course [7], diseases [8], germi-nation [9], weed distinguishing proof [10], etc.Also the system designed by [11] offered higher accuracy based on optimiztion in real time. It used Fast R-CNN which adapts to complex and rapid flying environment changes.…”

Section: Related Workmentioning

confidence: 99%

“…The robotic stage rating [9,10] of maize grains is difficult, and the majority of the work is done manually. The majority of the maize harvest is utilized for livestock feed, public consumption, and a variety of commercial items.…”

Section: Theoretical Analysismentioning

confidence: 99%

Maize Grain Quality Classification Using Convolutional Neural Network

M¹,

Anitha²,

Revathi³

2021

Advances in Parallel Computing Technologies and Applications

View full text Add to dashboard Cite

The major goal of our research would be to classify the quality of maize grain after obtaining the crop output. Grains have been the most important grain for farmers to boost their farming revenue in a country. One of the most significant food crops is wheat grain. The majority of the country’s grain is used for animal feed, human consumption, and the creation of industrial products. The amount of mechanization used to assess grain quality is modest, and the majority of the job is done by hand. Every day, different prices are introduced in various marketplaces, and farmers face losses every year quality of service estimate and market rates. With our classification we can give accurate progress on the seed which will help farmer’s to sell with better price along with that this assumption can idealize him to give good nutritious fertilizer to the crop.

show abstract

Implemetation of image classification CNN using multi thread GPU

Cited by 13 publications

References 1 publication

Real-Time Identification of Cyanobacteria Blooms in Lakeshore Zone Using Camera and Semantic Segmentation: A Case Study of Lake Chaohu (Eastern China)

Real-Time Identification of Cyanobacteria Blooms in Lakeshore Zone Using Camera and Semantic Segmentation: A Case Study of Lake Chaohu (Eastern China)

Automated Agronomy: Evaluation of Fruits Ripeness Using Machine Learning Approach

Maize Grain Quality Classification Using Convolutional Neural Network

Contact Info

Product

Resources

About