WO2019056402A1 - 一种基于灰度状态序列的送料分选装置 - Google Patents

一种基于灰度状态序列的送料分选装置 Download PDF

Info

Publication number
WO2019056402A1
WO2019056402A1 PCT/CN2017/103845 CN2017103845W WO2019056402A1 WO 2019056402 A1 WO2019056402 A1 WO 2019056402A1 CN 2017103845 W CN2017103845 W CN 2017103845W WO 2019056402 A1 WO2019056402 A1 WO 2019056402A1
Authority
WO
WIPO (PCT)
Prior art keywords
discharge
gray
standard
state sequence
main controller
Prior art date
Application number
PCT/CN2017/103845
Other languages
English (en)
French (fr)
Inventor
杨欢
王顺
Original Assignee
南京创优科技有限责任公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京创优科技有限责任公司 filed Critical 南京创优科技有限责任公司
Priority to EP17926030.2A priority Critical patent/EP3677350A4/en
Publication of WO2019056402A1 publication Critical patent/WO2019056402A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/34—Sorting according to other particular properties
    • B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/36—Sorting apparatus characterised by the means used for distribution
    • B07C5/361—Processing or control devices therefor, e.g. escort memory
    • B07C5/362—Separating or distributor mechanisms

Definitions

  • the invention belongs to the technical field of automatic production, and particularly relates to a feeding sorting device based on a sequence of gray states.
  • Automatic feeding is the first step in the realization of production automation.
  • the efficient and reliable operation of the feeding device is a necessary condition for the successful completion of automated production.
  • the requirements of the automated production line for the feeder are to deliver the material to the loading station at a constant speed and a uniquely determined attitude, while the material is often disordered and may be contaminated with defects or other types of materials.
  • the feeding device needs to screen the raw material state to meet the requirements of the loading station; on the other hand, it is necessary to remove defects or foreign matter in the material.
  • the core of the goal of automatic feeding is material sorting technology.
  • Material sorting technology is a technology that sorts different products or different grades according to certain standards or requirements to meet different needs. Material sorting not only improves the quality and effectiveness of the product, but also adapts to the differences in the needs of different users, enabling rational, economical and efficient use of resources, reducing environmental pollution and ecological damage. In this era of rapid economic development and diversified demand, resources are gradually consumed and environmental problems are becoming more and more serious. Material sorting is the main link in material circulation processing, which is very important from the perspectives of economics and ecology. significance.
  • Sorting the sorting method from the sorting mechanism mainly including screening, magnetic separation, light selection, electric selection, floating And other sorting methods, wherein the light sorting is a sorting method related to the optical characteristics of the sorting object, and the sorting materials are identified by optical characteristics such as absorption, reflection, and scattering of the sorting object.
  • the core of the optical sorting device is a set of photodetection devices and corresponding data processing systems.
  • the photodetecting device in the light sorting method generally has a light emitting unit and a light receiving unit.
  • the main form of the light emitting unit includes a point, line or surface light source composed of laser, infrared, ultraviolet and other visible light of various wavelengths.
  • the light receiving unit mainly covers a point gradation sensor, a line array charge coupler, a grayscale or color image sensor, and the like.
  • Chinese Patent Application (Publication No.: CN102989693B) discloses a material sorting method and device based on laser transmission, which uses a laser line source to illuminate the sorting materials, and transmits light according to the difference in transparency and volume scattering of different materials.
  • the back surface forms different intensity distributions, and the control system processes the intensity distribution signal and compares it with corresponding parameters of the material to be retained reserved in the system to determine whether it is a foreign object.
  • the method is suitable for sorting materials with good transparency, such as raisins, vegetables, fruits, seafood, etc., but it is difficult to achieve sorting for opaque objects.
  • Chinese patent application (Publication No.: CN106269570A) discloses a material sorting device and a sorting method for extracting features of sorting materials by double-infrared images of different wavelengths. The method distinguishes different materials according to the difference of infrared absorption of the substance. Compared with the infrared image of a single band, the method obtains more information from the two-color infrared image, which is beneficial to improve the correct rate of material sorting. Although this method is suitable for most materials with infrared absorption differences, there is a shortage of slow sorting speed and high system cost.
  • Chinese patent application (Publication No.: CN104390987A) discloses a novel optical fiber sensor and a detection method for detecting surface defects of a steel ball.
  • the optical fiber sensor uses 19 optical fibers to form a sensor probe, and solves the steel ball through 6 photoelectric conversion signals. Surface defects.
  • the method can accurately determine the roughness and displacement variation defects of the steel ball surface, but it is difficult to generalize to the material sorting system because it uses a large number of optical fibers and requires 6 signal conversion and collection processing operations.
  • Time series is a set of numerical sequences arranged in chronological order.
  • Time series analysis is a statistical method of dynamic data. Based on stochastic process theory and mathematical statistics method, this method studies the statistical laws followed by random data sequences to solve practical problems.
  • Time series analysis is based on the number of time series observed by the system According to the theory and method of establishing mathematical models through curve fitting and parameter estimation. Time series analysis is often applied in national economic macro control, regional comprehensive development planning, enterprise management, market potential prediction, meteorological hydrological forecast, earthquake pre-earthquake forecast, crop pest and disease forecast, environmental pollution control, ecological balance, astronomy and oceanography And other fields.
  • the present invention aims to provide a feeding sorting device based on a gray state sequence, which overcomes the defects of the prior art and realizes high speed and low by detecting a gray state sequence of materials. Material sorting with false positive rate.
  • a feeding sorting device based on a gray state sequence comprising a gray scale sensor, a main controller, a rejecting mechanism, a tray and a guiding trough; all materials are placed on the tray and advanced with the tray, and the gray scale sensor collects the material
  • the gray state sequence is transmitted to the main controller, and the main controller recognizes the gray state sequence;
  • the main controller matches the gray state sequence with the preset standard allowable discharge attitude data and each standard forbidden discharge attitude data, if the gray state sequence and the preset standard allow the discharge attitude data Matching, indicating that the material is in the target discharge attitude, allowing the material to pass through the reject mechanism; if the gray state sequence matches the preset standard prohibition discharge attitude data, it indicates that the material is in a non-discharge posture, the main controller
  • the control rejection mechanism returns the material to the initial position of the tray, and the posture of the returned material is changed by the falling process in the rejection mechanism and the vibration of the tray, and the system will perform the identification again when the rejection mechanism is re-arrived;
  • the main controller performs corresponding operations on the materials in this case according to the preset processing countermeasures;
  • the processing countermeasures include allowing the rejecting mechanism, prohibiting the passing of the rejecting mechanism, and stopping the alarm;
  • the final material passes the guiding trough to the discharging station in the target discharging attitude, or the operation is completed according to the preset processing countermeasures of the user.
  • each preset standard allows the discharge attitude data or the standard prohibition discharge posture data to include a plurality of gray states and a state transition sequence between the respective gray states, when the grayscale sensor collects the gray of the material
  • the sequence of states satisfies the state-shift sequence of the preset standard allowable discharge attitude data or the standard prohibition discharge attitude data, and determines that the gray state sequence of the material matches the standard allowable discharge attitude data or the standard forbidden discharge attitude data.
  • the preset standard allows the gray state of the discharge attitude data or the standard prohibition discharge attitude data to include a range of values of the gray value, and the frequency or duration of occurrence of the gray value.
  • the preset standard allows the discharge attitude data or the standard prohibition discharge posture data to be learned in advance by the feeding sorting device, so that the determined posture of the material advances on the tray, and the grayscale sensor collects the material.
  • the gray state sequence in the entire detection area is transmitted to the main controller, and the main controller analyzes its characteristics to obtain the gray state and the state transition order of the material, which is the preset standard attitude data, and is saved in the main control.
  • the standard attitude data is set by the user as standard allowable discharge attitude data or standard prohibition discharge attitude data.
  • the gradation sensor is a photoelectric sensor, and includes a light emitting unit that emits a modulated optical signal, and a light receiving unit that outputs a gradation value of the material at the position of the detected spot.
  • the photoelectric sensor adopts a through-beam photoelectric sensor, and if the material is non-transparent, the photoelectric sensor adopts a reflective photoelectric sensor.
  • the wavelength of the emitted light of the light emitting unit is adjusted according to the reflection characteristic of the material.
  • the main controller presets a plurality of standards to allow the discharge attitude data, the standard prohibition discharge posture data, and the countermeasures for the unmatched materials, and flexibly change the material discharge posture according to different requirements of the discharge station. A combination of discharge attitudes.
  • the gray state sequence of the important or confusing material is selected in advance as the feature template, and the operations of the corresponding culling mechanisms of each feature template are set, thereby facilitating distinguishing the state with high similarity; for the remaining ash not belonging to the feature template
  • the degree sequence is set in advance, and the processing of the corresponding culling mechanism or the alarm operation is performed according to the processing countermeasure, thereby reducing the number and difficulty of setting the feature template of the material that is easy to distinguish.
  • the present invention applies a time series analysis method to a material sorting system.
  • the calculation unit performs discrete time sampling on the gray scale sensor to obtain a gray state sequence, and analyzes and calculates the change of the state sequence.
  • the feature obtains category information, and the method can greatly improve the correct rate of material feature recognition compared to the single gray value classification method;
  • the invention does not need to stay in the photoelectric detection area during the feeding process, and the gray state sequence feature recognition method is faster than the image processing method, and is beneficial to improving the operation efficiency of the feeding device;
  • the invention is compatible with the on-beam and reflective gray scale sensors, and can meet the requirements of feeding sorting of transparent, translucent and non-transparent materials.
  • Figure 1 is a schematic view of the structure of the present invention
  • Figure 2 is a gray state transition diagram of the material discharge posture in the embodiment
  • Fig. 3 is a schematic diagram showing the gradation sequence of the different postures of the material A in the embodiment.
  • the invention designs a feeding sorting device based on a gray state sequence, as shown in FIG. 1 , which comprises three parts of a gray scale sensor 3 , a main controller 4 and a rejecting mechanism 5 , and combines the tray 1 and the guiding trough 2 .
  • a discharge station 6 constitutes a programmable universal feed sorting device.
  • the working process of the device is that all the materials are placed on the tray 1 and advance with the tray 1.
  • the gray scale sensor 3 collects the gray state sequence of the material and transmits it to the main controller 4, and the main controller 4 judges the collected gray scale. Whether the status sequence has been identified. If the gray state sequence has been identified, the main controller 4 matches the gray state sequence of the collected material with the preset standard discharge posture data, if the gray state sequence of the material and the preset standard discharge If the posture data matches, it indicates that the material is in the target discharge posture 7, and the material is allowed to pass through the reject mechanism 5.
  • the main controller 4 controls the reject mechanism 5 to return the material to the initial position of the tray 1, and the discharge posture of the returned material is changed by the drop of the material in the reject mechanism 5 and the vibration of the tray 1. If the gray state sequence is not recognized, the main controller 4 controls the tray 1 to stop running, and sends an abnormal alarm message to the worker. With the operation of the feed sorting device, all materials are finally passed through the guide trough 2 to the discharge station 6 in the target discharge attitude.
  • the gradation sensor 3 is a photosensor including a light emitting unit that emits a modulated optical signal, and a light receiving unit that outputs a gradation value of the material at the position of the detected spot. If the material is transparent or translucent, the photoelectric sensor adopts a through-beam photoelectric sensor, and if the material is non-transparent, the photoelectric sensor adopts a reflective photoelectric sensor. Moreover, when the photoelectric sensor adopts the reflective photoelectric sensor, the wavelength of the emitted light of the light emitting unit is adjusted according to the reflection characteristic of the material, so that the quality of the reflected signal is better and the signal to noise ratio is higher.
  • the standard discharge attitude data is a sequence of states consisting of a transition relationship between a gray state and a state corresponding to the material discharge posture.
  • the sequence of states is pre-designed in the main controller, and the transition conditions between all state transition sequences and states constitute a criterion for determining the type of material.
  • the sequence of states can be manually controlled to move the material at a suitable speed in front of the grayscale sensor at a suitable speed, and control the main controller to perform sampling learning. Therefore, the main controller is required to have a learning control interface, such as a learning button. After pressing the learning button, the main controller enters the learning state, and the correct material posture passes through the gray level sensor at a constant speed.
  • the main controller detects the state transition of the obtained gray scale sequence after the material completely passes the detection area. Data points with similar gray values are determined to be in the same gray state, so that a gray object containing the target material can be obtained.
  • a target feature can be defined as a sequence of states that contain a specific set of state transitions in that state space. Use S to indicate the grayscale state, G to use the grayscale value, and F to indicate the grayscale occurrence or the grayscale duration.
  • a specific preset target posture data is taken as an example to illustrate a specific implementation manner. A plurality of preset target posture data can be promoted according to the example. Such promotion does not involve creative labor, and therefore belongs to the scope of patent protection of the present invention.
  • the single target material discharge attitude state transition diagram is shown in Figure 2.
  • the condition for entering the S1 state is that the gray value is between 0.7 and 0.9 and the frequency at which the gray value continuously appears is equal to four.
  • the condition for entering the S2 state is that a gray sequence of three consecutive gray values in the range of 0.3 to 0.5 appears without a timeout.
  • the condition for entering S4 is that the gray value drops to less than 0.1 without a timeout.
  • the unconditional output of S4 is true, and the unconditional output of S3 is false.
  • Figure 3 shows four possible feed attitudes a, b, c, d of material A of Figure 1, from which it can be seen that the discharge station requirements for the discharge attitude are the pose a in Figure 3.
  • the main controller needs to calculate the four gray scale sequences in Figure 3, sort the a pose and return the materials in the b, c, and d poses to the tray through the reject mechanism.
  • the gradation sequence x is represented by the symbol G[x].
  • G[a] ⁇ 0.05,0.88,0.87,0.85,0.86,0.66,0.41,0.42,0.39,0.04 ⁇
  • G[b] ⁇ 0.03,0.38,0.41,0.43,0.40,0.39,0.61,0.78,0.86,0.01 ⁇
  • G[c] ⁇ 0.02,0.81,0.83,0.79,0.85,0.79,0.81,0.78,0.77,0.78,0.87,0.79,0.81,0.83,0.82,0.84,0.81,0.83,0.78,0.75,0.81,0.01 ⁇
  • G[d] ⁇ 0.04,0.15,0.43,0.51,0.81,0.82,0.79,0.81,0.81,0.78,0.79,0.84,0.79,0.85,0.81,0.79,0.83,0.81,0.35,0.15,0.01 ⁇
  • state transition determination is performed on the four gray scale sequences, and the result is:
  • the result of the state transition judgment on G[a] is S[a], and the output result is true.
  • the gray sequence represented by G[b] is empty because it does not satisfy the entry condition of S1, so the result is empty.
  • the gradation sequence indicated by G[c] can smoothly enter the S1 state, but waits for a timeout in the S1 state, thereby entering S3, and the result is false.
  • the judgment result of G[d] is the same as G[c].
  • the true representation is the same as the target discharge posture, and the false representation is different. Therefore, it can be concluded that the a state belongs to the target discharge attitude, and b, c, and d do not belong.
  • the material with abnormal discharge posture can be strictly screened out by the transition condition between the state sequence and the state.
  • the main controller can process the grayscale data received by the grayscale sensor in real time without waiting for the grayscale sequence acquisition of the entire material to be completed.
  • the grayscale state sequence analysis is a near real-time online data processing method. The whole gray scale sampling, feature recognition, calculation and judgment, and the response process are small, and can be completed before the material moves to the rejecting device. Such a design has almost no influence on the discharge speed.
  • the material type changes it can be re-learned to establish a new target material discharge attitude gray state transition diagram, which can guide the calculation unit to complete the conversion work with little manual intervention.
  • the invention provides a feeding sorting device based on a gray state sequence, which provides a new idea for solving the problem of automatic feeding.
  • Automatic feeding is an important part of automated production. Due to the wide variety of material objects and the wide variety of characteristics, the traditional feeding device is difficult to generalize, resulting in a higher price of the customized feeding device. Therefore, it is of great economic value and practical significance to realize a high efficiency and high reliability programmable universal feeding device.

Landscapes

  • Sorting Of Articles (AREA)

Abstract

一种基于灰度状态序列的送料分选装置，包括灰度传感器（3）、主控制器（4）、剔除机构（5）、料盘（1）和导料槽（2）；灰度传感器采集物料的灰度状态序列并传送给主控制器，主控制器将采集的物料的灰度状态序列与预设的标准出料姿态数据进行匹配，若该物料的灰度状态序列与预设的标准出料姿态数据相匹配，则表明该物料处于目标出料姿态（7），允许该物料通过剔除机构，否则表明该物料处于非出料姿态（8），主控制器控制剔除机构将该物料退回料盘的初始位置，通过料盘的振动，使退回的物料的出料姿态发生改变，随着送料分选装置的运行，最终物料均以目标出料姿态经过导料槽到达出料工位（6）。该装置通过检测物料的灰度状态序列，实现了高速且低误判率的物料分选。

Description

一种基于灰度状态序列的送料分选装置 技术领域
本发明属于自动化生产技术领域，特别涉及了一种基于灰度状态序列的送料分选装置。
背景技术
自动送料是生产自动化实现的第一个环节，送料装置的高效可靠运行是自动化生产顺利完成的必要条件。自动化生产线对送料装置的要求是以恒定的速度和唯一确定的姿态将物料送至上料工位，而原料通常是无序的且可能夹杂着缺陷或其他种类的物料。送料装置一方面需要对原料状态进行筛选以符合上料工位的要求；另一方面还需要剔除物料中的缺陷或异物。实现自动送料这一目标的核心就是物料分选技术。
物料分选技术是一种按照一定的标准或要求将不同的产品或不同等级的产品分类挑选以满足不同需求的技术。进行物料分选不但可以提升产品的品质和效用，而且还可以适应不同用户的需求差异，实现资源合理、经济、有效地利用，减少对环境的污染和对生态的破坏。在这个经济飞速发展、需求多样化细分的时代，资源逐渐消耗、环境问题日益严重，物料分选作为物料流通处理中的主要环节，从经济学和生态学的观点来看都具有十分重要的意义。
随着工业4.0、中国制造2025等概念的兴起，工业自动化蓬勃发展起来，物料分选技术也从低效的人工分选逐步发展到自动化分选。自动化分选技术相比以往的人工分选在效率、正确率等方面具有无可比拟的优势，所以其发展速度很快。由于需要进行分选的物料种类、外形等各种理化性质具有很大的差异，这也导致了分选过程的机理千差万别，由此发展出了一门专门的学科——物料分选学。
从分选机理方面对分选方法进行分类，主要有筛选、磁选、光选、电选、浮 选和其他分选方法，其中光分选是一种与分选对象光学特性相关的分选方法，通过分选对象对光的吸收、反射、散射等光学特性来识别分选物料。光分选装置的核心是一套光电检测装置与对应的数据处理系统。光分选方法中的光电检测装置通常都具有光发射单元和光接收单元。光发射单元主要的形式包括由激光、红外、紫外和其他各种波长的可见光构成的点式、线式或面光源。光接收单元主要涵盖点式灰度传感器、线阵电荷耦合器、灰度或彩色图像传感器等。
中国专利申请(公开号：CN102989693B)公开了一种基于激光透射的物料分选方法和装置，利用激光线光源对分选物料进行照射，根据不同物料的透明特性和体散射差异使透射光在其背面形成不同的强度分布，控制系统对该强度分布信号进行处理并与系统中预留的待保留物料的对应参数进行比对来判断是否是异物。该方法适用于透射性很好的物料的分选，如葡萄干、蔬菜、水果、海产品等，但对于不透明物体则较难实现分选。中国专利申请(公开号：CN106269570A)公开了一种物料的分选装置和分选方法，通过不同波长的双红外图像对分选物料的特征进行提取。该方法根据物质的红外吸收差异区分不同的物料，相对于单一波段的红外图像，该方法从双色红外图像中获得的信息更多，有利于提高物料分选的正确率。虽然该方法对于大多数具有红外吸收差异的物料适用，但存在分选速度慢、系统成本高的不足。中国专利申请(公开号：CN104390987A)公开了一种检测钢球表面缺陷的新型光纤传感器及检测方法，该光纤传感器使用了19根光纤构成一个传感器探头，通过6路光电转换信号解算钢球的表面缺陷。该方法可以准确测定钢球表面的粗糙度和位移变化缺陷，但由于其使用光纤数量较多且需要6路信号转换与采集处理操作，故较难推广至物料分选系统。
时间序列是按时间顺序排列的一组数字序列，时间序列分析是一种动态数据的统计方法，该方法基于随机过程理论和数理统计方法，研究随机数据序列所遵从的统计规律来解决实际问题。时间序列分析是根据系统观测得到的时间序列数 据，通过曲线拟合和参数估计来建立数学模型的理论和方法。时间序列分析常应用在国民经济宏观控制、区域综合发展规划、企业经营管理、市场潜力预测、气象水文预报、地震震前预报、农作物病虫灾害预报、环境污染控制、生态平衡、天文学和海洋学等领域。
总结现有的分选方法可以发现，为了实现较高的分选正确率常使用基于图像的方法，但受图像处理速度的限制无法实现高速分选；而为了实现高速分选常使用点式光电传感器来提高分选速度，但分选误判率较高。因此，需要一种能实现高速和低误判率的物料分选方法，物料灰度状态序列分析是实现该目标的一条技术途径。
发明内容
为了解决上述背景技术提出的技术问题，本发明旨在提供一种基于灰度状态序列的送料分选装置，克服了现有技术存在的缺陷，通过检测物料的灰度状态序列，实现高速且低误判率的物料分选。
为了实现上述技术目的，本发明的技术方案为：
一种基于灰度状态序列的送料分选装置，包括灰度传感器、主控制器、剔除机构、料盘和导料槽；所有物料置于料盘上并随料盘前进，灰度传感器采集物料的灰度状态序列并传送给主控制器，主控制器对此灰度状态序列进行辨识；
主控制器将该灰度状态序列与预设的各标准允许出料姿态数据和各标准禁止出料姿态数据进行匹配，若该灰度状态序列与预设的任一标准允许出料姿态数据相匹配，则表明该物料处于目标出料姿态，允许该物料通过剔除机构；若该灰度状态序列与预设的标准禁止出料姿态数据匹配，则表明该物料处于非出料姿态，主控制器控制剔除机构将该物料退回料盘的初始位置，通过剔除机构中的跌落过程和料盘的振动使退回的物料的姿态发生改变，重新到达剔除机构时系统会再次进行鉴别；
若该灰度状态序列与任一种标准允许出料姿态数据和标准禁止出料姿态数 据都不能匹配时，主控制器根据预先设置的处理对策，对这种情况的物料进行相应操作；所述处理对策包括允许通过剔除机构、禁止通过剔除机构和停机报警；
随着送料分选装置的运行，最终物料以目标出料姿态经过导料槽到达出料工位，或按照用户预设的处理对策完成操作。
进一步地，每个预设的标准允许出料姿态数据或标准禁止出料姿态数据均包括多个灰度状态以及各个灰度状态之间的状态转移顺序，当灰度传感器采集到物料的灰度状态序列满足预设的标准允许出料姿态数据或标准禁止出料姿态数据中的状态转移顺序，则判定该物料的灰度状态序列与标准允许出料姿态数据或标准禁止出料姿态数据相匹配。
进一步地，预设的标准允许出料姿态数据或标准禁止出料姿态数据中的灰度状态包含灰度值的取值范围，以及该灰度值出现的频数或持续的时间。
进一步地，所述预设的标准允许出料姿态数据或标准禁止出料姿态数据是预先通过送料分选装置学习得到的，令物料以确定的姿态在料盘上前进，灰度传感器采集该物料在整个检测区域内的灰度状态序列并传送给主控制器，主控制器分析其特征得到该物料包含的灰度状态以及状态转移顺序，即为预设的标准姿态数据，并保存在主控制器内部，该标准姿态数据由用户设定作为标准允许出料姿态数据或标准禁止出料姿态数据。
进一步地，所述灰度传感器为光电传感器，包括光发射单元和光接收单元，光发射单元发射经过调制的光信号，而光接收单元输出物料在检测光斑位置处的灰度值。
进一步地，若物料为透明或半透明时，所述光电传感器采用对射式光电传感器，若物料为非透明时，所述光电传感器采用反射式光电传感器。
进一步地，当光电传感器采用反射式光电传感器时，根据物料的反射特性调节光发射单元的发射光波长。
进一步地，主控制器中预设多个标准允许出料姿态数据、标准禁止出料姿态数据以及对不匹配物料的处理对策，根据出料工位的不同要求灵活改变物料的出料姿态或多种出料姿态的组合。
进一步地，预先选择重要或易混淆的物料的灰度状态序列作为特征模板，并设置各特征模板各自对应的剔除机构的操作，从而便于区分相似度高的状态；对于不属于特征模板的其余灰度状态序列，预先设定处理对策，根据处理对策进行对应的剔除机构的操作或报警操作，从而降低容易区分状态的物料的特征模板设置数量和难度。
采用上述技术方案带来的有益效果：
(1)本发明将时间序列分析方法应用于物料分选系统中，在物料经过光电检测区域时计算单元对灰度传感器进行离散时间采样获得一个灰度状态序列，通过分析计算该状态序列的变化特征得到类别信息，该方法相比于单灰度值的分类方法，可以极大提高物料特征识别的正确率；
(2)本发明在送料过程中无需在光电检测区域停留，且灰度状态序列特征识别方法相比图像处理方法速度更快，均有利于提高送料装置的运行效率；
(3)本发明兼容对射式和反射式灰度传感器，可以满足透明、半透明和非透明物料的送料分选需求。
附图说明
图1是本发明的结构示意图；
图2是实施例中物料出料姿态灰度状态转移图；
图3是实施例中物料A不同姿态的灰度序列示意图。
附图标记说明：1、料盘；2、导料槽；3、灰度传感器；4、主控制器；5、剔除机构；6、出料工位；7、目标出料姿态；8、非出料姿态。
具体实施方式
以下将结合附图，对本发明的技术方案进行详细说明。
本发明设计了一种基于灰度状态序列的送料分选装置，如图1所示，包括灰度传感器3、主控制器4和剔除机构5三部分，并结合料盘1、导料槽2和出料工位6组成一套可编程通用的送料分选装置。
该装置的工作过程为，所有物料置于料盘1上并随料盘1前进，灰度传感器3采集物料的灰度状态序列并传送给主控制器4，主控制器4判断采集的灰度状态序列是否已被识别。若该灰度状态序列已被识别，主控制器4将采集的物料的灰度状态序列与预设的标准出料姿态数据进行匹配，若该物料的灰度状态序列与预设的标准出料姿态数据相匹配，则表明该物料处于目标出料姿态7，允许该物料通过剔除机构5，若该物料的灰度状态序列与预设的标准出料姿态数据不匹配，则表明该物料处于非出料姿态8，主控制器4控制剔除机构5将该物料退回料盘1的初始位置，通过物料在剔除机构5中的跌落以及料盘1的振动，使退回的物料的出料姿态发生改变；若该灰度状态序列未被识别，则主控制器4控制料盘1停止运行，同时向工作人员发出异常报警信息。随着送料分选装置的运行，最终所有物料均以目标出料姿态经过导料槽2到达出料工位6。
在本实施例中，灰度传感器3为光电传感器，包括光发射单元和光接收单元，光发射单元发射经过调制的光信号，而光接收单元输出物料在检测光斑位置处的灰度值。若物料为透明或半透明时，所述光电传感器采用对射式光电传感器，若物料为非透明时，所述光电传感器采用反射式光电传感器。且当光电传感器采用反射式光电传感器时，根据物料的反射特性调节光发射单元的发射光波长，使反射信号质量更好、信噪比更高。
为了进行灰度状态匹配操作，需要预先定义标准出料姿态数据。标准出料姿态数据是由与物料出料姿态对应的灰度状态及状态之间的转移关系构成的状态序列。该状态序列是在主控制器中预先设计好的，所有状态转移顺序和状态之间的转移条件构成判断物料类型的标准。该状态序列可以通过人工控制将物料以目标姿态在灰度传感器前以合适的速度匀速运动，同时控制主控制器进行采样学习构造而成。因此，需要主控制器具备学习控制接口，例如学习按键。在按下学习按键后主控制器进入学习状态，将正确的物料姿态在灰度传感器前匀速通过，主控制器检测到物料完全通过检测区域后对获得的灰度序列进行状态转换。灰度值相近的数据点判定为同一灰度状态，因此可以得到一个包含目标物料姿态全部灰 度状态的状态空间。目标特征就可以定义为在该状态空间中包含一组特定状态转移顺序的状态序列。使用S表示灰度状态，使用G表示灰度值，使用F表示灰度出现的频数或灰度持续时间。这里以单一预设目标姿态数据为例说明具体实施方式，多个预设目标姿态数据可以根据该实例进行推广，这样的推广并没有付出创造性的劳动，所以也属于本发明专利保护的范畴。
单一目标物料出料姿态状态转移图如图2所示。进入S1状态的条件是灰度值在0.7至0.9之间且灰度值连续出现的频数等于4。进入S2状态的条件是在没有超时的前提下出现连续3次灰度值在0.3至0.5区间的灰度序列。进入S4的条件是在没有超时的前提下灰度值下降到比0.1小。S4无条件输出结果为真，S3无条件输出结果为假。从图中可以看出，目标物料姿态灰度状态序列为M＝{S1->S2->S4}，只有与该状态序列相同的物料灰度状态序列才给出结果为真的判断，继而将该物料姿态判定为出料姿态，否则剔除机构执行动作将物料退回料盘。
下面以物料A的四个面为例说明灰度状态序列的推导过程。图3表示图1中物料A的四种可能的送料姿态a、b、c、d，从图1中可以看出出料工位对出料姿态的要求是图3中的姿态a。主控制器需要计算图3中的四个灰度序列，将a姿态分选出来并将b、c、d姿态的物料通过剔除机构退回料盘。使用符号G[x]表示灰度序列x。通过测量得到：
G[a]＝{0.05,0.88,0.87,0.85,0.86,0.66,0.41,0.42,0.39,0.04}
G[b]＝{0.03,0.38,0.41,0.43,0.40,0.39,0.61,0.78,0.86,0.01}
G[c]＝{0.02,0.81,0.83,0.79,0.85,0.79,0.81,0.78,0.77,0.78,0.87,0.79,0.81,0.83,0.82,0.84,0.81,0.83,0.78,0.75,0.81,0.01}
G[d]＝{0.04,0.15,0.43,0.51,0.81,0.82,0.79,0.81,0.81,0.78,0.79,0.84,0.79,0.85,0.81,0.79,0.83,0.81,0.35,0.15,0.01}
根据图2中各状态的转移条件，对这四个灰度序列进行状态转移判断，结果为：
S[a]＝{S1->S2->S4}
S[b]＝{}
S[c]＝{S1->S3}
S[d]＝{S1->S3}
对G[a]进行状态转移判断的结果为S[a]，输出结果为真。G[b]表示的灰度序列由于不满足S1的进入条件，其状态序列为空，所以结果为空。G[c]表示的灰度序列可以顺利进入S1状态，但是在S1状态等待超时，从而进入S3，结果为假。G[d]的判断结果与G[c]相同。
通过对序列状态结果转移图中输出结果的判断，真表示与目标出料姿态相同，假表示不同。所以可以得出结论：a状态属于目标出料姿态，b、c、d均不属于。
从上面的判断结果可以看出通过状态序列及状态之间的转移条件可以严格将出料姿态异常的物料筛选出来。主控制器可以实时处理灰度传感器接收的灰度数据，无需等待整个物料灰度序列采集完成，灰度状态序列分析是一种近实时的在线数据处理方法。整个灰度采样、特征识别、计算判断、做出响应的过程计算负担小、在物料运动到剔除装置前便可完成，这样的设计对出料速度几乎没有影响。在物料类型改变时可以重新学习，建立新的目标物料出料姿态灰度状态转移图，只需很少的人工介入便可以指导计算单元完成这一转换工作。
本发明提供的一种基于灰度状态序列的送料分选装置，为解决自动化送料问题提供了一种新思路。自动送料是自动化生产中十分重要的环节，由于物料对象种类繁多、特征千差万别，传统的送料装置很难实现通用化，导致了定制送料装置价格较高。因此，实现一种高效率、高可靠的可编程通用送料装置具有十分重要的经济价值和现实意义。
以上实施例仅为说明本发明的技术思想，不能以此限定本发明的保护范围，凡是按照本发明提出的技术思想，在技术方案基础上所做的任何改动，均落入本发明保护范围之内。

Claims (9)

  1. 一种基于灰度状态序列的送料分选装置，其特征在于：包括灰度传感器、主控制器、剔除机构、料盘和导料槽；所有物料置于料盘上并随料盘前进，灰度传感器采集物料的灰度状态序列并传送给主控制器，主控制器对此灰度状态序列进行辨识；
    主控制器将该灰度状态序列与预设的各标准允许出料姿态数据和各标准禁止出料姿态数据进行匹配，若该灰度状态序列与预设的任一标准允许出料姿态数据相匹配，则表明该物料处于目标出料姿态，允许该物料通过剔除机构；若该灰度状态序列与预设的标准禁止出料姿态数据匹配，则表明该物料处于非出料姿态，主控制器控制剔除机构将该物料退回料盘的初始位置，通过剔除机构中的跌落过程和料盘的振动使退回的物料的姿态发生改变，重新到达剔除机构时系统会再次进行鉴别；
    若该灰度状态序列与任一种标准允许出料姿态数据和标准禁止出料姿态数据都不能匹配时，主控制器根据预先设置的处理对策，对这种情况的物料进行相应操作；所述处理对策包括允许通过剔除机构、禁止通过剔除机构和停机报警；
    随着送料分选装置的运行，最终物料以目标出料姿态经过导料槽到达出料工位，或按照用户预设的处理对策完成操作。
  2. 根据权利要求1所述基于灰度状态序列的送料分选装置，其特征在于：每个预设的标准允许出料姿态数据或标准禁止出料姿态数据均包括多个灰度状态以及各个灰度状态之间的状态转移顺序，当灰度传感器采集到物料的灰度状态序列满足预设的标准允许出料姿态数据或标准禁止出料姿态数据中的状态转移顺序，则判定该物料的灰度状态序列与标准允许出料姿态数据或标准禁止出料姿态数据相匹配。
  3. 根据权利要求2所述基于灰度状态序列的送料分选装置，其特征在于： 预设的标准允许出料姿态数据或标准禁止出料姿态数据中的灰度状态包含灰度值的取值范围，以及该灰度值出现的频数或持续的时间。
  4. 根据权利要求2所述基于灰度状态序列的送料分选装置，其特征在于：所述预设的标准允许出料姿态数据或标准禁止出料姿态数据是预先通过送料分选装置学习得到的，令物料以确定的姿态在料盘上前进，灰度传感器采集该物料在整个检测区域内的灰度状态序列并传送给主控制器，主控制器分析其特征得到该物料包含的灰度状态以及状态转移顺序，即为预设的标准姿态数据，并保存在主控制器内部，该标准姿态数据由用户设定作为标准允许出料姿态数据或标准禁止出料姿态数据。
  5. 根据权利要求1所述基于灰度状态序列的送料分选装置，其特征在于：所述灰度传感器为光电传感器，包括光发射单元和光接收单元，光发射单元发射经过调制的光信号，而光接收单元输出物料在检测光斑位置处的灰度值。
  6. 根据权利要求5所述基于灰度状态序列的送料分选装置，其特征在于：若物料为透明或半透明时，所述光电传感器采用对射式光电传感器，若物料为非透明时，所述光电传感器采用反射式光电传感器。
  7. 根据权利要求6所述基于灰度状态序列的送料分选装置，其特征在于：当光电传感器采用反射式光电传感器时，根据物料的反射特性调节光发射单元的发射光波长。
  8. 根据权利要求1所述基于灰度状态序列的送料分选装置，其特征在于：主控制器中预设多个标准允许出料姿态数据、标准禁止出料姿态数据以及对不匹配物料的处理对策，根据出料工位的不同要求灵活改变物料的出料姿态或多种出料姿态的组合。
  9. 根据权利要求1所述基于灰度状态序列的送料分选装置，其特征在于：预先选择重要或易混淆的物料的灰度状态序列作为特征模板，并设置各特征模板 各自对应的剔除机构的操作；对于不属于特征模板的其余灰度状态序列，预先设定处理对策，根据处理对策进行对应的剔除机构的操作或报警操作。
PCT/CN2017/103845 2017-09-20 2017-09-28 一种基于灰度状态序列的送料分选装置 WO2019056402A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP17926030.2A EP3677350A4 (en) 2017-09-20 2017-09-28 INPUT SORTING DEVICE BASED ON GRAYSCALE STATE SEQUENCE

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710853450.9 2017-09-20
CN201710853450.9A CN107671011B (zh) 2017-09-20 2017-09-20 一种基于灰度状态序列的送料分选装置

Publications (1)

Publication Number Publication Date
WO2019056402A1 true WO2019056402A1 (zh) 2019-03-28

Family

ID=61136629

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/103845 WO2019056402A1 (zh) 2017-09-20 2017-09-28 一种基于灰度状态序列的送料分选装置

Country Status (3)

Country Link
EP (1) EP3677350A4 (zh)
CN (1) CN107671011B (zh)
WO (1) WO2019056402A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114951047A (zh) * 2022-05-26 2022-08-30 河海大学 基于光纤传感器的振动送料中通用型智能分选方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109761017B (zh) * 2019-01-29 2024-05-03 无锡先导智能装备股份有限公司 一种入料口检测装置

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101111502B1 (ko) * 2011-03-11 2012-02-22 주식회사코마스 곡물 색채 선별기를 위한 향상된 동시 선별방법
CN102989693A (zh) 2012-12-28 2013-03-27 合肥美亚光电技术股份有限公司 一种基于激光透射的物料分选方法和装置
CN103008259A (zh) * 2012-12-28 2013-04-03 合肥美亚光电技术股份有限公司 基于激光透射的物料分选装置
CN203484357U (zh) * 2013-10-08 2014-03-19 合肥美亚光电技术股份有限公司 一种双红外在线塑料材质分选装置
CN104390987A (zh) 2014-12-16 2015-03-04 济南大学 一种检测钢球表面缺陷的新型光纤传感器及检测方法
CN105251707A (zh) * 2015-11-26 2016-01-20 长沙理工大学 基于医药大输液可见异物检测系统的次品剔除分拣装置
CN205165215U (zh) * 2015-11-24 2016-04-20 河南理工大学 一种干法块煤分选装置
CN106269570A (zh) 2015-05-29 2017-01-04 合肥美亚光电技术股份有限公司 物料的分选装置和分选方法

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3319243B2 (ja) * 1995-08-29 2002-08-26 神鋼電機株式会社 部品整送装置
CN102663362B (zh) * 2012-04-09 2014-11-05 宁波中科集成电路设计中心有限公司 一种基于灰度特征的运动目标检测方法
CN103101741B (zh) * 2013-02-04 2017-04-05 江苏中科百易矿业科技有限公司 激光视觉输送带损伤检测装置的实现方法
CN103896025B (zh) * 2014-04-02 2016-03-02 东南大学 基于智能相机的柔性振动传输系统及其工作方法
CN203997897U (zh) * 2014-07-04 2014-12-10 佛山科学技术学院 用于将马赛克颗粒顺序排列输出的供料装置
CN104773446B (zh) * 2015-03-26 2017-02-22 宁波科瑞特光电技术有限公司 用于补强钢片的振动盘供料系统
CN106980816B (zh) * 2017-02-22 2020-08-25 贵州电网有限责任公司凯里供电局 基于光学图像的绝缘子串自动识别方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101111502B1 (ko) * 2011-03-11 2012-02-22 주식회사코마스 곡물 색채 선별기를 위한 향상된 동시 선별방법
CN102989693A (zh) 2012-12-28 2013-03-27 合肥美亚光电技术股份有限公司 一种基于激光透射的物料分选方法和装置
CN103008259A (zh) * 2012-12-28 2013-04-03 合肥美亚光电技术股份有限公司 基于激光透射的物料分选装置
CN203484357U (zh) * 2013-10-08 2014-03-19 合肥美亚光电技术股份有限公司 一种双红外在线塑料材质分选装置
CN104390987A (zh) 2014-12-16 2015-03-04 济南大学 一种检测钢球表面缺陷的新型光纤传感器及检测方法
CN106269570A (zh) 2015-05-29 2017-01-04 合肥美亚光电技术股份有限公司 物料的分选装置和分选方法
CN205165215U (zh) * 2015-11-24 2016-04-20 河南理工大学 一种干法块煤分选装置
CN105251707A (zh) * 2015-11-26 2016-01-20 长沙理工大学 基于医药大输液可见异物检测系统的次品剔除分拣装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114951047A (zh) * 2022-05-26 2022-08-30 河海大学 基于光纤传感器的振动送料中通用型智能分选方法
CN114951047B (zh) * 2022-05-26 2023-08-22 河海大学 基于光纤传感器的振动送料中通用型智能分选方法

Also Published As

Publication number Publication date
EP3677350A4 (en) 2020-11-25
CN107671011B (zh) 2019-07-12
EP3677350A1 (en) 2020-07-08
CN107671011A (zh) 2018-02-09

Similar Documents

Publication Publication Date Title
EP0738186B1 (en) Method and apparatus for detecting surface features of translucent objects
CN106733701B (zh) 马铃薯多级联合智能缺陷检测系统及检测方法
US4378494A (en) Apparatus and method for detecting defects in glass bottles using event proximity
CN106409711A (zh) 一种太阳能硅晶片缺陷检测系统及方法
CN106093066A (zh) 一种基于改进的机器视觉注意机制的磁瓦表面缺陷检测方法
CN101105459A (zh) 一种空瓶瓶口缺陷的检测方法及装置
CN101915767B (zh) 一种鸡蛋裂纹的系统性识别方法
Zulkifley et al. Robust identification of polyethylene terephthalate (PET) plastics through bayesian decision
CN110756457A (zh) 墙地砖自动识别分类系统及方法
WO2008004559A1 (fr) système de groupage, et dispositif d'évaluation de type de défaut
CN103439337A (zh) 压电陶瓷蜂鸣片外观缺陷自动检测装置及方法
CN113145492A (zh) 一种用于梨外观品质视觉分级方法及分级生产线
CN111652883B (zh) 基于深度学习的玻璃表面缺陷检测方法
WO2019056402A1 (zh) 一种基于灰度状态序列的送料分选装置
CN104318249A (zh) 基于多分类支持向量机的磨削表面粗糙度等级识别方法
CN104132945A (zh) 一种基于光纤传导的棒材表面质量在线视觉检测装置
Alikhanov et al. Design and performance of an automatic egg sorting system based on computer vision
CN107891012B (zh) 基于等效算法的珍珠大小及圆形度分拣装置
EP1554060B1 (en) A system and a method of automatically sorting objects
JP6771190B2 (ja) 樹脂判定方法及び装置
WO2020190169A1 (ru) Способ сортировки объектов по их цветовым характеристикам
US4760270A (en) Method of and apparatus for comparing data signals in an electrooptical inspection system
CN110458231B (zh) 一种陶瓷产品检测方法、装置和设备
KR20130019818A (ko) 가시광 대역 플라스틱 판별 장치 및 이를 이용한 플라스틱 분류 시스템
CN109829496B (zh) 一种物理测量分类方法及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17926030

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017926030

Country of ref document: EP

Effective date: 20200331