Patents

::: Current Position: 🏠Homepage > Awards > Patents

Patents	Type	Certificate number
情緒模型之建立方法及以該情緒模型進行的情緒偵測方法	ROC (Taiwan)	I501093
疲勞駕駛警示裝置及疲勞駕駛警示方法	ROC (Taiwan)	I557002
嬰兒哭鬧含意判讀模型的建立方法及使用方法	ROC (Taiwan)	I571862
步態分析方法與系統	ROC (Taiwan)	I657800
LED發光元件之主動功因修正驅動系統	ROC (Taiwan)	I665935
口鼻異物遮蔽檢測方法及其系統	ROC (Taiwan)	I684959
臉部對稱性檢測方法及其系統	ROC (Taiwan)	I689285
一種聽音辨位拍照及後製智能剪裁之系統	ROC (Taiwan)	I691211
浮標定位監測方法及浮標定位監測系統	ROC (Taiwan)	I698656
主動式打火機聲源辨識系統	ROC (Taiwan)	I699543
應用於半特徵視覺式同步定位與建圖方法的半特徵視覺里程計	ROC (Taiwan)	I712993
姿態影像訊號與運動訊號之訊號匹配系統以及訊號匹配方法	ROC (Taiwan)	I722674
三維點雲資料處理法	ROC (Taiwan)	I731604
自由空間偵測系統及自由空間偵測方法	ROC (Taiwan)	I733449
非接觸式呼吸檢測方法及其系統	ROC (Taiwan)	I741708
嬰兒哭聲辨識修正方法及其系統	ROC (Taiwan)	I747392
光源光紋辨識系統及其方法	ROC (Taiwan)	I781410
金屬製品表面之文字辨識方法	ROC (Taiwan)	I786946
一種實體物件轉換點雲資訊之模型建置系統及其方法	ROC (Taiwan)	I805934
LED顯示器控制方法	ROC (Taiwan)	I777379
虛實整合建模系統及方法	ROC (Taiwan)	I794621
生理訊號量測方法及其系統	ROC (Taiwan)	I816602
基於地板限制重定位與八叉樹優化建圖的三維光達式同步定位建圖及導航方法	ROC (Taiwan)	I856544
基於混合模糊之模糊控制的防車體側翻方法及其模糊控制系統	ROC (Taiwan)	I851104
利用光達式與視角式視覺伺服技術的上架對位移動控制方法	ROC (Taiwan)	I867464
圖面之文字及圖元辨識系統及方法	ROC (Taiwan)	I863841
洗車台監控系統及其監控方法	ROC (Taiwan)	I862228
支援車聯網應用服務的行車監控系統	ROC (Taiwan)	M545723
心率檢測裝置	ROC (Taiwan)	M590433
一種實體物件轉換雲資訊之模型建置系統	ROC (Taiwan)	M608782
高光譜影像之自動化咖啡豆瑕疵檢測方法及其系統	ROC (Taiwan)	M624851
洗車台監控系統	ROC (Taiwan)	M653952
HEART RATE DETECTION METHOD AND DEVICE THEREOF	USA	US10,722,126B2
FACIAL STROKING DETECTION METHOD AND SYSTEM THEREOF	USA	US10,846,518B2
MOUTH AND NOSE OCCLUDED DETECTING METHOD AND SYSTEM THEREOF	USA	US10,891,845B2
INERTIAL POSITIONING NAVIGATION DEVICE A NOVEL DETECTION METHOD	USA	US10,533,874B2
LED PANEL CONTROLLING METHOD AND LED PANEL CONTROLLING SYSTEM	USA	US11,107,393B1
METHOD FOR PROCESSING THREE-DIMENSIONAL POINT CLOUD DATA	USA	US11,348,261B2
METHOD AND SYSTEM FOR CORRECTING INFANT CRYING IDENTIFICATION	USA	US11,380,348B2
NAVIGATION SYSTEM APPLYING AUGMENTED REALITY	USA	US9,953,461


Name of Technology	Application of Hyperspectral Smart Agriculture in Food Safety Taking Eggs as Examples
Developer of Technology	Department of Computer Science and Information Engineering Shih-Yu Chen, Professor
Summary of Technology	This technology uses Snapshot to take a hyperspectral image of egg defects and uses a line scan camera to take a picture of egg freshness. The obtained hyperspectral image is processed and trained using an AI depth learning algorithm. In terms of prediction speed, this technology can achieve real-time identification (3.23 ms/image), and the overall accuracy of the two classifications can reach more than 95%. Technical characteristic: 1. Non-invasive: Using the non-invasive hyperspectral camera 2. Real-time: Conducting real-time analysis by AI depth learning algorithm. Accuracy can reach more than 95% and the prediction time of one image only takes 3.23ms. 3. Economize manpower: Saving the manpower of traditional manual sections. 4. Increase production: Long-running


Name of Technology	Using Hyperspectral Technology to Estimate Vase Life of Anthurium Cut
Developer of Technology	Department of Computer Science and Information Engineering Shih-Yu Chen, Professor
Summary of Technology	The use of hyperspectral information with multi-band and non-interaction characteristics combined with machine learning methods to estimate the vase life of anthurium and accurately predict the aging time of anthurium to increase market value.
Video


Name of Technology	Smart New Agriculture Real-time Automated Hyperspectral Defect Detection System
Developer of Technology	Department of Computer Science and Information Engineering Shih-Yu Chen, Professor
Summary of Technology	After using the snapshot-type hyperspectral camera to obtain the image of coffee beans, the recognition accuracy can be greatly improved through the band selection and AI deep learning technology, and the automatic integration with the robot arm can accurately and instantly screen the defective coffee beans It turns out that this technology is not only a fast, comprehensive and non-invasive detection method, it can improve coffee quality and reduce a lot of labor costs. This technology can be applied to the screening of defects in various agricultural products, such as coffee, peanuts, and nuts.
Video


Name of Technology	Smart Product Grading and Defect Screening Peanuts
Developer of Technology	Department of Computer Science and Information Engineering Shih-Yu Chen, Professor
Video


Name of Technology	Health Education Kebbi Air Robot
Developer of Technology	Department of Computer Science and Information Engineering Yi-Lung Lin, Associate Professor
Video


Name of Technology	Automatic Detection Technology for IR Thermal Defects in Solar Modules
Developer of Technology	Director, IRIS Research Center Department of Computer Science and Information Engineering Chuan-Yu Chang, Distinguished Professor
Summary of Technology	Any defects in the solar cells, such as cracks, poor welding, or improper connections, can cause increased resistance, which, over time, leads to the formation of hot spots. The long-term effects of hot spots include burn marks, which reduce the efficiency of the solar cells and the backsheet, accelerate battery aging, and if left unchecked, could eventually lead to fires. The AI model developed by our team can mark and train known hot spot locations in IR images, enabling the detection of hot spots in IR video footage.
Video


Name of Technology	Automatic Detection Technology for EL Inspection in Solar Modules
Developer of Technology	Director, IRIS Research Center Department of Computer Science and Information Engineering Chuan-Yu Chang, Distinguished Professor
Summary of Technology	This technology has novel technologies such as adaptable and adjustable Yolo architecture, incremental learning and transfer learning, which can accurately detect dozens of defects in real-time with accuracy rate of 99.8%. This technology has been certified by KIWA in the Netherlands and has created several world firsts, making it a leader in solar module defect detection technology. This technology can be applied to any solar module, ranging from single-chip cells to large-scale defect detection of module arrays.
Video


Name of Technology	Detect Defects in Abrasive Wheel
Developer of Technology	Director, IRIS Research Center Department of Computer Science and Information Engineering Chuan-Yu Chang, Distinguished Professor
Summary of Technology	The black color of the abrasive wheel makes it difficult to visually identify defects, often leading to over-detection or missed defects. Our team effectively detects shape defects such as deformations, surface irregularities, chipped edges, and cracks. Additionally, our solution overcomes challenges related to the reflective metallic surface of the wheel and the limited availability of defect samples, providing enterprises with a comprehensive AI-driven inspection solution.


Name of Technology	Character Recognition and Defect Detection of Metal
Developer of Technology	Department of Computer Science and Information Engineering Yi-Lung Lin, Associate Professor
Summary of Technology	Casting Since the metal surface is easy to reflect light, it is not easy to identify using traditional optical character detection. This technology is used to identify the characters of several English/number symbols on the control valve. The symbols include numbers 0 to 9 and English letters A to Z. The technology that has been used includes image preprocessing, deep convolution layer, time series layer, and connectionist temporal classification layer. Considering the field application, it is designed as a lightweight model running on the mobile phone, and a set of character optical detection systems that integrate computer vision. This technology reduces the identification time and improves the accuracy of identification, which is convenient for subsequent application in product inventory management and production history filing after production is completed. Technical characteristic: 1. Digital management of products traceability 2. Reducing the working hours for field personnel to input data 3. Increasing the efficiency of the assembly and shipping process 4. Improving the production history system

Patents

Intelligent Recognition Industry Service Research Center

No. 123, Section 3, University Road, Douliu City, Yunlin County (3F-AI304/AI305,Dream Hub Building) Tel: 05-5342601 ext. 2776-2780 05-5524539

© 2020 IRIS YunTech All Rights Reserved

No. 123, Section 3, University Road, Douliu City, Yunlin County (3F-AI304/AI305,Dream Hub Building)
Tel: 05-5342601 ext. 2776-2780　05-5524539