This paper presents an innovative approach to automatic volume control using imageprocessing and deeplearning techniques. The ability to automatically adjust volume levels based on environmental factors and user pre...
详细信息
In this study, the development of a robotic cell that combines deeplearning and imageprocessing hybrid approach has been addressed in order to increase the accuracy and efficiency of the quality control of automotiv...
详细信息
In this study, the development of a robotic cell that combines deeplearning and imageprocessing hybrid approach has been addressed in order to increase the accuracy and efficiency of the quality control of automotive parts. In the automotive industry, manual quality control processes performed by operators are susceptible to errors and inaccuracies, leading to the passage of faulty parts and subsequent inefficiencies, wasted time, and increased costs. To overcome these challenges, this study introduces a fault detection robotic cell that combines deeplearning and imageprocessing techniques for quality control of automotive parts at Sahinkul Machine Spare Parts Manuf. Ltd. Co.. The robotic cell uses imageprocessing to inspect geometric tolerances, including hole diameter, part geometry and the presence of holes. However, the complex geometry of bolt threads requires the use of the YOLOv5 deeplearning algorithm to assess their quality. A dataset consisting of 3500 bolt thread images was collected for training and validation, with 2800 images used for training, 350 for validation, and the remaining 350 for testing purposes. The experimental results show that the fault detection robotic workcell achieves an approximate success rate of 97.4% in inspecting the quality of the selected parts. By combining deeplearning and imageprocessing, this study provides a reliable solution to improve the accuracy and efficiency of quality control processes in the automotive industry.
Press brakes are among the most important machines used in sheet metal processing. In these machines, different numbers of molds are used for sheet bending and these molds are placed in the system by an operator. Howe...
详细信息
Press brakes are among the most important machines used in sheet metal processing. In these machines, different numbers of molds are used for sheet bending and these molds are placed in the system by an operator. However, this process is slow, error-prone, and dependent on human labor. In this study, a real-time system that automatically detects molds and manipulates a robotic arm was designed using YOLOv4 and imageprocessing. YOLOv4, a deeplearning (DL)-based object detection algorithm, was applied to detect the positions, types, and holes of molds. Classical imageprocessing methods were implemented to find the center (X, Y) coordinates of the mold hole. This study shows that the press brake machines currently used in industry can be transformed into smart machines through DL, imageprocessing, camera systems, and robotic arm features.
In today's world, technology is changing our way of life and work at an alarming rate. This paper studies the performance of an improved deeplearning algorithm in imageprocessing tasks, introduces the implementa...
详细信息
Compared to the artificial identification of deterioration modes in stone cultural heritage, machine identification is more objective, detailed, accurate, timely, and cost-effective. Hyperspectral data can expand the ...
详细信息
Compared to the artificial identification of deterioration modes in stone cultural heritage, machine identification is more objective, detailed, accurate, timely, and cost-effective. Hyperspectral data can expand the information dimension and improve machine recognition capabilities. However, this augmentation of information introduces challenges in recognition efficiency, storage, and transmission. To address these challenges, this paper presents a deeplearning radial basis function (RBF) compression algorithm, aimed at enhancing the efficacy of hyper- spectral image analysis. The experimental results show that the identification model's F1-score remained around 0.95, with an average improvement in identification accuracy of 1.4%. Overall identification efficiency was enhanced by 13.8%, the identification model's training time was reduced by an average of 4.7%, and the identification time was reduced by an average of 9.1%. It provides a new scheme based on hyperspectral for nondestructive testing of stone cultural relics. And provides the corresponding business support for the relevant units.
Urban flooding is escalating worldwide due to the increasing impervious surfaces from urban developments and frequency of extreme rainfall events by climate change. Traditional flood extent prediction and mapping meth...
详细信息
Urban flooding is escalating worldwide due to the increasing impervious surfaces from urban developments and frequency of extreme rainfall events by climate change. Traditional flood extent prediction and mapping methods based on physical-based hydrological principles often face limitations due to model complexity and computational burden. In response to these challenges, there has been a notable shift toward satellite imageprocessing and Artificial Intelligence (AI) based approaches, such as deeplearning (DL) models, including architectures like Convolutional Neural Networks (CNN). The objective of this research is to predict near real-time (NRT) flood extents within urban areas. This research integrated CNN (U-Net) with Sentinel-1 satellite imagery, Digital Elevation Model (DEM), hydrologic soil group (HSG), imperviousness, and rainfall data to create a flood extent prediction model. To detect flooded areas, a binary raster map was created using calibrated backscatter values derived from the VV (vertical transmit and vertical receive) polarization mode of Sentinel-1 imagery, which was highlighted as having a significant impact on backscatter behavior and prediction results. Application of the model was demonstrated in urban areas of Miami-Dade County, Florida. The results demonstrated the capability of the model to provide rapid and accurate flood extent predictions at a spatial resolution of 10 m, with an overall accuracy of 97.05 %, F-1 Score of 92.49 %, and AUC of 93 % in the study area. The U-Net model's flood predictions were compared with historical floodplain data and then using GIS overlay analysis, resulting in a Ground Truth Index of 84.05 % that shows the accuracy of the model in identifying flooded areas. The research incorporated crucial flood-influencing data (including rainfall) to the flood extent prediction models and expanded the focus models beyond major rainfall events only to encompass a wider range of flood events. The presented NRT flood e
The proceedings contain 20 papers. The topics discussed include: a review of real-time human action recognition involving vision sensing;embedded real-time people detection and tracking with time-of-flight camera;real...
ISBN:
(纸本)9781510643093
The proceedings contain 20 papers. The topics discussed include: a review of real-time human action recognition involving vision sensing;embedded real-time people detection and tracking with time-of-flight camera;real-time object detection in 360-degree videos;a real-time software framework for driver monitoring systems: software architecture and use cases;realtime circle detection by simplified Hough transform on smartphones;the evaluation of CUDA performance on the Jetson NanoBoard for an image binarization task;chest X-ray classification using transfer learning on multi GPU;and parallel implementation of a hyperspectral feature extraction method based on Gabor filter.
To achieve target detection and defect recognition in power inspection images, an imageprocessing and recognition algorithm based on deeplearning is proposed. This algorithm mainly adopts an improved Faster-RCNN mod...
详细信息
Tuberculosis (TB) has been known as the top two murdering disease worldwide after HIV/AIDS. Indonesia is the second country with the highest number of tuberculosis. One of the reasons for the high number of cases is t...
详细信息
Sludge morphology plays a vital role in activated sludge (AS) settleability. However, the presence of filamentous bacteria has been a significant issue that degrades the performance of AS systems. In this study, a cos...
详细信息
Sludge morphology plays a vital role in activated sludge (AS) settleability. However, the presence of filamentous bacteria has been a significant issue that degrades the performance of AS systems. In this study, a cost-effective setup with an automated stage was used to capture AS floc images from three wastewater treatment plants in Sapporo, Japan, from October 2021 to September 2022. The AS floc parameters, such as aspect ratio, roundness, equivalent diameter, and the area of filamentous and weak flocs (A F &Wfs-QIA ), were quantified using imageprocessing software. The A F &Wfs-QIA parameter served as a quantitative measure of filaments in the software- based approach. Indications of filamentous bulking in spring and winter were observed by analyzing the correlation between the sludge volume index (SVI) and the morphological parameters. This study successfully demonstrates that with an increase in the fraction of small and less dense flocs, the SVI elevates. Furthermore, sludge with excess filamentous bacteria made supernatants cleaner in the settling tests. These results led to a new hypothesis involving a suspended-solid-trapping mechanism with filamentous bacteria. In addition, this study extends its scope to the segmentation and detection of areas covered by filaments through a U-Net deep-learning multi-label semantic segmentation technique. In the U-Net-based method, the area of filaments (A F-UNet ) was used to quantify protruding filaments. It achieved performance comparable to that in the literature with a minimal training dataset while reducing the processingtime. Both the software-based and U-Net-based approaches were evaluated against SVI to assess the impact of filament growth on sludge settleability.
暂无评论