Improving Defect Inspection Quality of Deep-Learning Network in Dense Beans by Using Hough Circle Transform for Coffee Industry

Kuo, Chung‐Hao; Chen, Chao‐Chun; Wang, Ding-Chau; Chen, Tzu-Ting; Chou, Yung-Chien; Pai, Mao-Yuan; Horng, Gwo-Jiun; Hung, Min‐Hsiung; Lin, Yu-Chuan; Hsu, Tz-Heng

doi:10.1109/smc.2019.8914175

Cited by 10 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain the best result, the inverse ratio of the resolution was set to 1, the upper threshold for the internal Canny edge detector was set to 10, the threshold for center detection was set to 9, and the range of radius to be detected was (20, 60) pixel. The ROI was extracted from the seven dye region labeled by the Hough circle transform …”

Section: Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

Integrated Portable Shrimp-Freshness Prediction Platform Based on Ice-Templated Metal–Organic Framework Colorimetric Combinatorics and Deep Convolutional Neural Networks

Zhang

et al. 2021

ACS Sustainable Chem. Eng.

View full text Add to dashboard Cite

Real-time monitoring of food freshness is critical to reducing food waste and pursuing sustainable development. Cross-reactive artificial scent screening systems provide a promising solution for food freshness monitoring, but their commercialization is hindered by the low sensitivity or pattern-recognition inaccuracy. Leveraging the cutting-edge artificial intelligence and high-porosity nanomaterial, a cost-effective and versatile method was developed by incorporating metal–organic frameworks into smart food packaging via a colorimetric combinatorics sensor array. The whole UiO-66 family was screened by density functional theory calculations, and UiO-66-Br (due to the highest binding energy) was selected to construct sensor arrays on an ice-templated chitosan substrate (i.e., ice-templated dye@UiO-66-Br/Chitosan). The physicochemical properties and morphologies of the fabricated sensor arrays were systematically characterized. The limit of detection of 37.17, 25.90, and 40.65 ppm for ammonia, methylamine, and trimethylamine, respectively, was achieved by the prepared composite film. Deep convolutional neural networks (DCNN), a deep learning algorithm family, were further applied to monitor shrimp freshness by recognizing the scent fingerprint. Four state-of-the-art DCNN models were trained using 31,584 labeled images and 13,537 images for testing. The highest accuracy achieved was up to 99.94% by the Wide-Slice Residual Network 50 (WISeR50). Our newly developed platform is integrated, sensitive, and non-destructive, enabling consumers to monitor shrimp freshness in real-time conveniently.

show abstract

Section: Machine Learningmentioning

confidence: 99%

“…The ROI was extracted from the seven dye region labeled by the Hough circle transform. 25 Database Preparation and Image Collection for DCNN Training. To collect data from the sensor, 10 s of a short video for each box was taken with shrimps at a certain storage interval.…”

Section: ■ Machine Learningmentioning

confidence: 99%

Integrated Portable Shrimp-Freshness Prediction Platform Based on Ice-Templated Metal–Organic Framework Colorimetric Combinatorics and Deep Convolutional Neural Networks

Zhang

et al. 2021

ACS Sustainable Chem. Eng.

View full text Add to dashboard Cite

show abstract

“…Thus, our defect-sensitive inspection scheme needs to meticulously differentiate defective beans from normal ones. Figure 5 shows the architecture of the DIDN, which is an improvement version of our previously designed deep network [36]. The DIDN consists of four main components and each of them are described below.…”

Section: Design Of Defect-sensitive Inspection Deep Network (Didn)mentioning

confidence: 99%

“…Our proposed scheme is implemented with mixture of Tensorflow, Python, and related tools, for generating the optimal DIDN model M DS * on a desktop computer with Intel Core i5-8500 3.0 GHz, 24 GB RAM, and a NVIDIA GTX 2080 Ti GPU card. For comparison, we also implemented two defect inspection schemes with the HCADIS [36] and the YOLOv3 [24,37]. Figure 11 shows experimental equipment, which is a robotic arm with a camera on the end effector for removing defective coffee beans used in this case study (Authors from seven Taiwan universities share system development load and accomplish experiments in the case study).…”

Section: Experimental Settings and Performance Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Deep-Learning-Based Defective Bean Inspection with GAN-Structured Automated Labeled Data Augmentation in Coffee Industry

et al. 2019

Self Cite

View full text Add to dashboard Cite

In the production process from green beans to coffee bean packages, the defective bean removal (or in short, defect removal) is one of most labor-consuming stages, and many companies investigate the automation of this stage for minimizing human efforts. In this paper, we propose a deep-learning-based defective bean inspection scheme (DL-DBIS), together with a GAN (generative-adversarial network)-structured automated labeled data augmentation method (GALDAM) for enhancing the proposed scheme, so that the automation degree of bean removal with robotic arms can be further improved for coffee industries. The proposed scheme is aimed at providing an effective model to a deep-learning-based object detection module for accurately identifying defects among dense beans. The proposed GALDAM can be used to greatly reduce labor costs, since the data labeling is the most labor-intensive work in this sort of solutions. Our proposed scheme brings two main impacts to intelligent agriculture. First, our proposed scheme is can be easily adopted by industries as human effort in labeling coffee beans are minimized. The users can easily customize their own defective bean model without spending a great amount of time on labeling small and dense objects. Second, our scheme can inspect all classes of defective beans categorized by the SCAA (Specialty Coffee Association of America) at the same time and can be easily extended if more classes of defective beans are added. These two advantages increase the degree of automation in the coffee industry. The prototype of the proposed scheme was developed for studying integrated tests. Testing results of a case study reveal that the proposed scheme can efficiently and effectively generate models for identifying defective beans with accuracy and precision values up to 80 % .

show abstract

Machine learning techniques for coffee classification: a comprehensive review of scientific research

Motta,

Vuillerme,

Pham

et al. 2024

Artif Intell Rev

View full text Add to dashboard Cite

In the realm of agribusiness, transformative shifts are underway, propelled by the growing demands and expanding scales of grain production. This evolution calls for a critical reevaluation of the existing paradigms in coffee production and marketing paradigms, with a specific focus on integrating Artificial Intelligence (AI). This work aims to review, synthesize, and summarize the available data regarding how Machine Learning (ML) has been used to detect and classify characteristics in coffee beans and leaves. For this purpose, a comprehensive literature review of the most significant research contributions describing the application of AI for advanced classification techniques in coffee agriculture has been carried out. Our analysis suggests that implementing AI technologies allows the classification of coffee, encompassing various attributes such as maturity, roast intensity, disease identification, flavor profiles, and overall quality. More largely, this technological advancement holds the potential to revolutionize coffee farming by providing producers and agricultural specialists with sophisticated tools to enhance production efficiency, minimize costs, and improve the accuracy and confidence of their decision-making processes. The motivation for the literature review is to address the increasing global demands and evolving scales of grain production, particularly in coffee farming, by critically reevaluating existing paradigms and integrating AI techniques. This review aims to synthesize and summarize how ML has been utilized to detect and classify various characteristics of coffee beans and leaves, thereby highlighting the potential of AI to revolutionize coffee farming by enhancing production efficiency, minimizing costs, and improving decisionmaking accuracy. This article presents the latest studies in ML in the coffee area, observes the methodology used, and allows researchers to develop new solutions that cover gaps in the literature, open problems, challenges, and future trends, bringing a real contribution to the scientific field. Finally, this article gathers and presents the databases used in many studies, which may be useful for future ML projects.

show abstract

Improving Defect Inspection Quality of Deep-Learning Network in Dense Beans by Using Hough Circle Transform for Coffee Industry

Cited by 10 publications

References 14 publications

Integrated Portable Shrimp-Freshness Prediction Platform Based on Ice-Templated Metal–Organic Framework Colorimetric Combinatorics and Deep Convolutional Neural Networks

Integrated Portable Shrimp-Freshness Prediction Platform Based on Ice-Templated Metal–Organic Framework Colorimetric Combinatorics and Deep Convolutional Neural Networks

Deep-Learning-Based Defective Bean Inspection with GAN-Structured Automated Labeled Data Augmentation in Coffee Industry

Machine learning techniques for coffee classification: a comprehensive review of scientific research

Contact Info

Product

Resources

About