会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明公开
    • 영상 처리 방법 및 장치
    • 图像处理方法及其设备
    • KR1020130044099A
    • 2013-05-02
    • KR1020110108378
    • 2011-10-21
    • 엘지이노텍 주식회사한양대학교 산학협력단
    • 박정아지호진김회율곽재호현종민나문수
    • G06T17/00G06T7/00G06T5/00
    • G06T7/0079G06T7/10G06T7/70G06T2207/10028
    • PURPOSE: An image processing method and an apparatus are provided to detect a target object simply and accurately by using only depth information. CONSTITUTION: An image processing device obtains the depth information of an image and separates a first ROI(Region Of Interest) of a target object existence region from an image(S11,S12). If a candidate region of the target object detected from the first ROI corresponds to the target object, an apparatus uses the candidate region in order to detect a target region(S13-S15). The apparatus estimates the posture of the target object by using the target region(S16). The apparatus models the target object(S17). [Reference numerals] (AA) Start; (BB) No; (CC) Yes; (DD) End; (S11) Obtain depth information of an image; (S12) Separate a region of interest(ROI); (S13) Detect a candidate region from the ROI; (S14) Candidate region matched with a target object?; (S15) Detect a target region using the candidate region; (S16) Estimate the posture of the target object; (S17) Model the target object
    • 目的:提供一种图像处理方法和装置,通过仅使用深度信息来简单且准确地检测目标对象。 构成:图像处理装置获取图像的深度信息,并将图像中的目标对象存在区域的第一ROI(感兴趣区域)分离(S11,S12)。 如果从第一ROI检测到的目标对象的候选区域对应于目标对象,则装置使用候选区域来检测目标区域(S13-S15)。 该装置使用目标区域来估计目标物体的姿势(S16)。 设备对目标物体进行建模(S17)。 (附图标记)(AA)开始; (BB)否 (CC)是; (DD)结束; (S11)获取图像的深度信息; (S12)分离感兴趣区域(ROI); (S13)从ROI检测候选区域; (S14)与目标对象相匹配的候选区域; (S15)使用候补区域检测目标区域; (S16)估计目标物体的姿势; (S17)模拟目标对象
    • 4. 发明公开
    • 식판 음식 자동 계량 및 계산 시스템
    • 食品检验方法及其设备
    • KR1020140094761A
    • 2014-07-31
    • KR1020130007102
    • 2013-01-22
    • 한양대학교 산학협력단
    • 김회율곽재호
    • G01N21/84G01B11/00G01N33/02G06T7/00
    • The objective of the present invention is to make users properly select food depending on the own amount of the food and to set a reasonable price according to the amount of the food. Thus, the present invention can protect domestic natural environment and can save resources by effectively reducing the amount of food wastes approaching 8,000 billion won per year. A restaurant needs fast and accurate calculation due to the constraints of time and space since there are a lot of people coming in all at the same time. To solve the problem, adopting a method using a computer vision is essential. The system divides the image of a food tray obtained by a distance camera for providing distance information and a color camera into each food section by analyzing the image of the food tray, and determines the kinds of the food in each divided section by matching the food with a today′s set menu. Then, the system measures the volume of the food tray using the distance camera to calculate the volume of the food. Accordingly, the system obtains weight information from the volume information on the food and calculates and displays the calorie and price of the food.
    • 本发明的目的是使用户根据食物的数量适当地选择食物,并根据食物量设定合理的价格。 因此,本发明可以保护国内自然环境,节约资源,有效减少每年接近8000亿韩元的食品废物。 由于时间和空间的限制,餐厅需要快速准确的计算,因为有很多人同时进入。 为了解决这个问题,采用计算机视觉的方法至关重要。 该系统通过分析食物托盘的图像将通过距离摄像机获得的食物托盘的图像分配给每个食物部分,并通过匹配食物来确定每个分割部分中的食物的种类 用今天的菜单。 然后,系统使用距离摄像机测量食物托盘的体积以计算食物的体积。 因此,系统从食物的体积信息中获取体重信息,并计算并显示食物的卡路里和价格。