WO2017193624A1 - 基于rfid的rssi信号值的图书放置姿态识别方法 - Google Patents

基于rfid的rssi信号值的图书放置姿态识别方法 Download PDF

Info

Publication number
WO2017193624A1
WO2017193624A1 PCT/CN2017/070930 CN2017070930W WO2017193624A1 WO 2017193624 A1 WO2017193624 A1 WO 2017193624A1 CN 2017070930 W CN2017070930 W CN 2017070930W WO 2017193624 A1 WO2017193624 A1 WO 2017193624A1
Authority
WO
WIPO (PCT)
Prior art keywords
book
rfid tag
rssi
time
rfid
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2017/070930
Other languages
English (en)
French (fr)
Inventor
陈力军
罗园洁
刘佳
万凌昊
陈曦
曾阿凡
李颖
施庆朴
黄嘉琪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University
Nanjing Tech University
Original Assignee
Nanjing University
Nanjing Tech University
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 Nanjing University, Nanjing Tech University filed Critical Nanjing University
Priority to US15/739,014 priority Critical patent/US10282576B2/en
Publication of WO2017193624A1 publication Critical patent/WO2017193624A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
    • G06K7/10366—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves the interrogation device being adapted for miscellaneous applications
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
    • G06K7/10366—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves the interrogation device being adapted for miscellaneous applications
    • G06K7/10376—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves the interrogation device being adapted for miscellaneous applications the interrogation device being adapted for being moveable
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management

Definitions

  • the invention relates to the field of RFID technology, in particular to a book placement gesture recognition method based on RFID RSSI signal values.
  • RFID Radio Frequency Identification
  • RFID is a non-contact automatic identification technology, also known as radio frequency identification (RFID), which is a communication technology that can identify specific targets and read and write related data through radio signals without identifying the system and Establish mechanical or optical contact between specific targets.
  • RFID technology has broad application prospects and is gradually being applied to all aspects of industrial production and daily life. Especially in the library, more and more libraries have begun to adopt RFID technology to replace the original barcode system.
  • the bar code system of the traditional library has the following disadvantages: 1.
  • the bar code adopts the laser technology. When it is necessary to scan a book information, it is necessary to rely on the artificial opening of the book, and scanning the bar code in the book with the scanner requires a lot of labor; 2.
  • the barcode can store less data, and the data is not encoded, the confidentiality and security of the data is not high; 3.
  • the barcode can only identify the producer and the product, and cannot identify the specific commodity.
  • Radio frequency identification (RFID) technology can solve these problems. Its outstanding advantages are as follows: 1. Non-contact reading: RFID tags can be read through non-metallic materials, and do not need to be in direct contact with the labels, so no manual Intervention to complete the identification work and achieve automation; 2.
  • RFID tag data storage capacity is large, the data on the tag can be encrypted, updated and readable and writable, especially suitable for storing large amounts of data or items. The stored data is often changed. 3.
  • Read and write speed RFID technology can recognize high-speed moving objects and recognize multiple labels at the same time, and the operation is quick and convenient.
  • High data security in addition to password protection, the data part is available. Some algorithms implement security management, such as DES, RSA, DSA, MD5, etc., and the reader and the electronic tag can also authenticate each other to achieve secure communication and storage. It can be said that the introduction of RFID technology has brought great convenience to library management. At present, the library has not used the RSSI signal of RFID tags for the recognition of book placement posture.
  • the books are occasionally placed on other books at will, resulting in disordered book discharge, affecting the management of library books, and can use RFID tag signals to automatically identify these wrong horizontals.
  • the book is convenient for the librarian to manage the books, which makes the library look neat and orderly, and also brings convenience to readers to find books.
  • the technical problem to be solved by the present invention is to provide a book placement gesture recognition method based on RFID-based RSSI signal value, which utilizes the RSSI value in the RFID technology to automatically identify a library that is incorrectly placed horizontally. Books on the shelf.
  • the present invention discloses a book placement gesture recognition method based on an RFID-based RSSI signal value, including: a method for recognizing a book placement gesture based on a change rule of an RFID tag RSSI value over time, and the specific steps are as follows:
  • Step 1 Using an RFID device to read an RFID tag placed in the book, the RFID tag includes: an EPC number epc of the book RFID tag, an RSSI value rssi of the RFID tag, a time t at which the RFID tag is read, and the above three pieces of information are represented as one
  • Step 2 classifying the triplets with the same EPC number into one class, as the RSSI signal value of the internal RFID tag acquired by the book at different times;
  • Step 3 preprocessing the triplet obtained in step 2, smoothing the data, and removing impurities;
  • Step 4 Establish a quadratic curve model for the preprocessed data in step 3, and obtain a variation rule of the RSSI value of the RFID tag of the book with time;
  • Step 5 Using the quadratic curve model established in step 4 to train, using Naive Bayes, SVM and other algorithms to obtain a classification model; collecting the RFID tag signal of the book to be identified, and determining the placement posture of the to-be-recognized book as vertical according to the classification model Placed on a bookshelf or lying flat on a bookshelf;
  • step 6 the ranking algorithm is used to finally confirm the position of the book, and it is determined whether the step 5 is determined to be incorrect.
  • step 2 includes:
  • the triad obtained in step 1 is classified according to the EPC number, and the triplet holding the same EPC number is divided into a class P, indicating the RSSI signal value read by a book at different times.
  • step 3 includes the following steps:
  • step 3-1 the triples in each type P in step 2 are sorted according to time, the triples with the small triplet time t are in front, and the triples with large time t are in the back;
  • Step 3-2 smoothing the data
  • Step 3-3 Obtain a maximum RSSI value in each type of P and a corresponding time t, which is marked as maxindex;
  • Step 3-4 for each type of P, traversing by the maxindex forward and backward triplets. If the time difference ⁇ t of the adjacent two triplets p is greater than a threshold, the triplets outside the triplet are eliminated. Data to remove more discrete data.
  • Step 3-5 for each type of P, traversal by the maxindex forward and backward triples:
  • the rssi value in the triple should be smaller and smaller, if traversing to the tern of the rssi value becomes larger
  • the triple data before the triple is removed; during the backward traversal, the rssi value in the triple should be smaller and smaller. If you traverse to the triple with the rssi value starting to increase, reject this. Triad data after the triple.
  • step 4 includes:
  • the quadratic curve fitting is performed on the pre-processed data in step 3, and the quadratic curve model is established to obtain the variation rule of the RSSI value of the book RFID tag with time.
  • the quadratic curve model is expressed as follows:
  • a, b, and c represent the coefficients of the quadratic curve, respectively.
  • the coefficients a, b, and c of the quadratic model are solved by the least squares method.
  • step 5 includes the following steps:
  • Step 5-1 randomly take out a certain number of books (generally about 5% to 30%) and place them on other vertically placed books, read the RFID tags of these books and perform pre-processing to establish a quadratic curve model;
  • Step 5-2 using a classification algorithm such as SVM classification algorithm and naive Bayesian classification algorithm to train the quadratic curve model established in step 5-1 to obtain a classification model;
  • a classification algorithm such as SVM classification algorithm and naive Bayesian classification algorithm
  • Step 5-3 after obtaining the classification model, reading the RFID tag of the book to be identified, performing preprocessing, establishing a quadratic curve model, placing the quadratic curve model into the classification model, performing classification and judging, and identifying the placement posture of the book as vertical Place it on the bookshelf or lie flat on the bookshelf.
  • step 6 includes the following steps:
  • Step 6-1 if it is determined in step 5-3 that the book is placed flat on the bookshelf, the RFID device is vertically moved to read the RFID tag signal placed in the book;
  • step 6-2 the RSSI book sorting algorithm is used to finally confirm the position of the book, and it is judged whether the result of step 5 is incorrect.
  • step 6-2 includes the following steps:
  • Step 6-2-1 classify the triples with the same EPC number into one class, perform data preprocessing and quadratic curve fitting;
  • Step 6-2-2 the time T of the peak of the fitted curve, that is, the highest point, is obtained, and the time T of all the RFID tags is sorted to obtain the position order of each book corresponding to the RFID tag;
  • Step 6-2-3 if the time of the RFID tag is arranged in order, and the time of the book that is determined to be lying on the bookshelf is ranked last, it indicates that the step 5 is correct, otherwise the step 5 is determined to be incorrect.
  • the book is placed perpendicular to the bookshelf.
  • the present invention utilizes the RSSI signal of the RFID tag signal for the recognition of the book placement posture, accurately determines the state in which the book is placed on the bookshelf, and detects the books lying on other books in the bookshelf, which facilitates the library management personnel to correct the book.
  • the wrong placement poses to keep the library books in a neat and orderly state.
  • Figure 1 is a top plan view showing the positional relationship between the antenna and the bookshelf.
  • Figure 2 is a front view of a book placed horizontally on a bookshelf and a book placed vertically on a bookshelf.
  • Figure 3 is a graph of RSSI values as a function of time.
  • Figure 4 is a flow chart of an embodiment.
  • Figure 5 is a flow chart of book sorting.
  • RFID technology consists of three parts: an antenna, a reader and an RFID tag.
  • the antenna queries the RFID tag by transmitting a radio frequency signal, and the radio frequency signal is reflected back by the tag after reaching the RFID tag, and the reader recognizes the RFID tag by receiving the reflected signal, and the RSSI signal is a kind of reflected signal.
  • the RSSI signal value reflected by the tag is used to realize the recognition of the library book placement posture.
  • the distance is related to the RSSI signal as follows:
  • RSSI is the received signal strength (negative value)
  • A is the signal strength when the transmitting end and the receiving end are separated by 1 meter
  • n is the environmental attenuation factor. It can be converted from the above formula:
  • the above formula indicates that the smaller the distance d is, the larger the RSSI value is. In this embodiment, the closer the distance between the antenna and the book RFID tag is, the larger the RSSI value is.
  • the antennas In the process of identifying the position and posture of the books on the library shelves, the antennas should be placed in parallel not far from the books, otherwise the RSSI value will not be very different for the size of a book at different times. , the establishment of the impact model.
  • the distance of the antenna from the book is set to 10 cm.
  • the antenna is slowly moved from one side of the bookshelf to the other side of the bookshelf at a uniform speed, as shown in FIG.
  • the RSSI value of the read RFID tag is increasing; when the antenna is away from the RFID tag, the RSSI value is decreasing, and the RSSI value read at the RFID tag is the largest;
  • the embodiment provides a book placement gesture recognition method based on the RFID technology RSSI signal value, wherein the data collection process is as follows:
  • the tag contains the epc number including the book id number.
  • the corresponding book id number is extracted from the book, and then from the background database server. Query the book information corresponding to the corresponding id number.
  • the information of the book is collected, including the horizontally placed book information and the vertically placed book information.
  • the quadratic curve model is classified by a naive Bayes algorithm and an SVM algorithm to construct a classification model.
  • the triples in each type of P are sorted according to time, the triplet p with a small triplet time t is in front, and the triplet p with a large time t is after.
  • the RSSI value reflected by the tag will cause interference, resulting in unsatisfactory data.
  • the data needs to be smoothed.
  • the value of RSSI in each triplet is taken as the RSSI value of the adjacent two triplets. Average value to reduce data jitter.
  • the point of the RSSI value that may appear at some time is far from the curve, and some discrete points or small waves appear, so these sparseness must be The point is processed. For each type of P, it is traversed by the maxindex forward and backward triplets. If the time difference ⁇ t of the adjacent two triplets p is greater than a threshold, the data of the triples outside the triplet is eliminated. Remove more discrete data.
  • the threshold is set according to the size of time. In this embodiment, the unit of time is seconds, so this threshold is set to 0.1 second.
  • the placement posture of the book is judged, and the book placement state needs to be finally confirmed by the book sorting algorithm.
  • the specific process is as follows:
  • Data classification, preprocessing, and the process of establishing a quadratic curve model are the same as the data processing process of book placement gesture recognition.
  • the library book placement gesture recognition process based on the RFID tag RSSI value of this embodiment is shown in FIG. 4 .
  • the tag epc number is bound to the book id number, and bind the book information to the background database server.
  • the format of the RFID tag epc number in the book is set to:
  • 0 indicates that the type of the tag is a book RFID tag.
  • XXXXX indicates the book ID number, obtained from the barcode of the book
  • the 8-digit number starting from the 4th digit of the epc number is extracted as the id number of the book, and the background database contains the id information of the book and the specific information of the book, such as the title of the book, the author, and the like.
  • the book information can be queried according to the id number to the background database.
  • the books on the library shelves should be neatly discharged, otherwise the effect of recognition will be affected. Place all the books After tidying, let the antennas be placed in parallel at a distance of about 10 cm from the book. The antenna can't be too far away from the book. Otherwise, the value of the RSSI signal returned by a book will not be much different during the process of moving, because there is no big change in the distance of the antenna relative to the book within a certain range.
  • the speed of movement should not be too fast, the moving speed is fast, and the number of times the antenna reads the label is reduced. In this embodiment, the speed is controlled to be about 0.1 m/s.
  • Data collection should be divided into two, one is to collect training data, to obtain the classification model and threshold, to prepare for the latter identification process, and the other is to collect the data of the book to identify the placement posture, so the corresponding data processing and model
  • the establishment process is also carried out twice.
  • the data recorded in the back-end database is usually a staggered arrangement of data of multiple books, and these interlaced data need to be classified to extract the corresponding triple set of each book.
  • the classification method used is to divide the data with the same epc number into one class, and then sort the data in the divided triple set according to the size of time. Due to environmental factors, the interference of the RSSI signal of the RFID tag is large, and the thickness of the book is different.
  • the RSSI signal data collected in each book is not ideal, and there is impurity data, which needs to be removed by preprocessing.
  • the ideal data arrangement should assume a shape that approximates a quadratic curve, so it is necessary to eliminate discrete points and small peaks.
  • the model should be a quadratic curve with an open downward. The larger the opening, the longer the read RFID tag is read, and the more reads, the more The greater the probability that the RFID tag is in a horizontal position.
  • the construction of the model is divided into two, one for the data training to obtain the threshold of the classification model and classification, and the other is the identification of the book placement posture, if the quadratic coefficient of the curve obtained in the recognition process and the threshold value of the classification model If the difference is within a certain small range, it can be judged that the book is in a horizontal position.
  • the judgment result of the above process may have a certain error due to the influence of environmental factors on the RSSI signal, resulting in a judgment error.
  • the final judgment In the judgment of the position where the book is placed horizontally, the book is further scanned vertically, data is collected, and the books are sorted. If there is only one vertically placed book in this position, then the final sorting result should have only one time, or several similar time values, then it proves that the judgment is wrong and the book is placed correctly. Indeed. If there is a book that is placed across other books in this position, the final sorting result should be the time of more than one book, and the time interval is not small, and the reading time of the book judged to be horizontally placed is behind. .
  • the flow of library book sorting based on the RFID tag RSSI value of this embodiment is as shown in FIG. 5.
  • the direction of the book sorting scan book is perpendicular to the direction of the scanned book in the book gesture recognition process, and is scanned upward from the bottom end of the book shelf to the upper end of the book shelf.
  • the rest of the steps are the same as the recognition process of the book placement gesture.
  • the acquired data is the data of multiple books interlaced, the data needs to be extracted and classified, and the data with the same epc number is grouped into one class, representing the RSSI value of the RSSI tag of a book returned at different times.
  • the data of the well-classed data is obtained in chronological order to obtain the RSSI value.
  • the graph obtained in the above steps will have discrete points, or some small peaks will appear. These are the interference factors for the subsequent steps, which need to be carried out. Eliminate, leaving only a continuous curve with the highest value.
  • the pre-processed curve is not the best model for analysis. It is necessary to perform quadratic curve fitting on the curve to obtain a complete band that increases first and then decreases with time.
  • the time at which the maximum value is taken is T. .
  • the obtained T 1 , T 2 , T 3 ... Tn are sorted in ascending order, which is the order of the books in which the tags are located.
  • the present invention provides a method for identifying a book placement gesture based on an RFID-based RSSI signal value.
  • the method and the method for implementing the technical solution are numerous.
  • the above description is only a preferred embodiment of the present invention, and it should be noted that the common technology in the technical field is Many modifications and refinements can be made without departing from the principles of the invention, and such modifications and refinements are also considered to be within the scope of the invention.
  • the components that are not clear in this embodiment can be implemented by the prior art.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Toxicology (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Electromagnetism (AREA)
  • General Health & Medical Sciences (AREA)
  • Business, Economics & Management (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种基于RFID的RSSI信号值的图书放置姿态识别方法，包括：步骤1，使用RFID设备横向移动读取放置在图书内的RFID标签信号；步骤2，将带有相同标签EPC号的三元组归为一类，作为一本图书在不同时刻获取到的其内部RFID标签的RSSI信号值；步骤3，对步骤2获取的数据进行预处理；步骤4，建立模型，得到图书RFID标签RSSI值随时间的变化规律；步骤5，采样数据，训练出分类模型，根据测试数据建立好的模型和分类模型，判断图书的放置姿态；步骤6，使用RFID设备竖向移动读取放置在图书内的RFID标签信号，利用排序算法最终确认图书的位置，判断步骤5结果是否有误。

Description

基于RFID的RSSI信号值的图书放置姿态识别方法 技术领域
本发明涉及一种RFID技术领域，特别是基于RFID的RSSI信号值的图书放置姿态识别方法。
背景技术
RFID(Radio Frequency Identification,射频识别)是一种非接触式的自动识别技术，又称无线射频识别，是一种通信技术，可通过无线电讯号识别特定目标并读写相关数据，而无需识别系统与特定目标之间建立机械或光学接触。RFID技术应用前景广，目前正在逐步地应用于工业生产和日常生活的各个方面，尤其在图书馆方面，越来越多的图书馆开始采用RFID技术来取代原有的条形码系统。
传统图书馆的条形码系统具有以下缺点：1.条形码采用激光技术，在需要扫描一本图书信息的时候，必须依靠人为的翻开书本，用扫描器扫描图书内的条形码，需要耗费大量的人工；2.条形码可存储的数据比较少，而且数据未经过编码，数据的保密性和安全性不高；3.条形码只能识别生产者和产品，并不能辨认具体的商品。无线射频识别(RFID)技术可以解决这些问题，其突出的优点主要有以下几点：1.非接触阅读：RFID标签可以透过非金属材料阅读，而且不需要与标签直接接触，因此能够无需人工干预地完成识别工作，实现自动化；2.数据存储容量大：RFID标签的数据存储容量大，标签上的数据可以加密、随时更新及可读可写，特别适合于存储大量数据或物品上所需存储的数据经常改变请情况；3.读写速度快：RFID技术可识别高速运动物体并可同时识别多个标签，操作快捷方便；4.数据安全性高：标签除了密码保护外，数据部分可用一些算法实现安全管理，如DES、RSA、DSA、MD5等，读写器与电子标签之间也可相互认证，实现安全通信和存储。可以说，RFID技术的引入给图书馆管理带来了极大的便利，目前图书馆还没有将RFID标签的RSSI信号用于图书放置姿态的识别。在图书馆中，由于部分读者的粗心，偶尔会将图书随意横放在其他图书上，造成了图书排放紊乱，影响了图书馆图书的管理，可以利用RFID标签信号，自动识别这些被错误的横放的图书，方便了图书管理员对图书的管理，使得图书馆呈现整洁有序的面貌，也给读者查找图书带来了便利。
发明内容
发明目的：本发明所要解决的技术问题是针对现有技术的不足，提供基于RFID的RSSI信号值的图书放置姿态识别方法，利用RFID技术中的RSSI值来自动识别图书馆中错误地横放于书架上的图书。
为了解决上述技术问题，本发明公开了基于RFID的RSSI信号值的图书放置姿态识别方法，包括：基于RFID标签RSSI值随时间的变化规律对图书放置姿态识别的方法，具体步骤如下：
步骤1，使用RFID设备读取放置在图书内的RFID标签，RFID标签包括：图书RFID标签的EPC号epc，RFID标签RSSI值rssi，读取RFID标签的时刻t，将以上三个信息表示为一个三元组p＝{epc,rssi,t}，代表一本图书在t时刻读取到的放置在其内部的RFID标签的RSSI值；
步骤2，将带有相同EPC号的三元组归为一类，作为一本图书在不同时刻获取到的内部RFID标签的RSSI信号值；
步骤3，对步骤2获取的三元组进行预处理，平滑数据，剔除杂质；
步骤4，对步骤3中预处理好的数据建立二次曲线模型，得到图书RFID标签RSSI值随时间的变化规律；
步骤5，利用步骤4建立的二次曲线模型进行训练，利用朴素贝叶斯、SVM等算法得出分类模型；采集待识别图书RFID标签信号，根据分类模型，判断待识别图书的放置姿态为垂直于书架放置或平躺于书架放置；
步骤6，利用排序算法最终确认图书的位置，判定出步骤5是否判断有误。
其中，步骤2包括：
将步骤1得到的三元组根据EPC号进行分类，将持有相同EPC号的三元组划分为一类P，表示一本图书在不同时刻读取到的RSSI信号值。
其中，步骤3包括如下步骤：
步骤3-1，对步骤2中每类P中的三元组根据时间排序，三元组时间t小的三元组在前，时间t大的三元组在后；
步骤3-2，对数据进行平滑处理；
步骤3-3，获取每类P中的最大RSSI值以及对应的时间t，标记为maxindex；
步骤3-4，对于每类P，由maxindex向前后的三元组遍历，如果相邻的两个三元组p的时间差Δt大于一个阈值，就剔除在此三元组之外的三元组的数据，以此去除较离散的数据。
步骤3-5，对于每类P，由maxindex向前后的三元组遍历：向前遍历过程中，三元组中的rssi值应该越来越小，如果遍历到rssi值开始变大的三元组，就剔除此三元组之前的三元组数据；向后遍历过程中，三元组中的rssi值也应该越来越小，如果遍历到rssi值开始变大的三元组，剔除此三元组之后的三元组数据。
其中，步骤4包括：
对步骤3中预处理好的数据进行二次曲线拟合，建立二次曲线模型，得到图书RFID标签RSSI值随时间的变化规律，二次曲线模型表示为如下形式：
rssi＝a×t2+b×t+c。
其中a，b，c分别表示二次曲线的系数。二次曲线模型的系数a，b，c采用最小二乘法求解。
其中，步骤5包括如下步骤：
步骤5-1，随机取出一定数量的图书(一般选取5％～30％左右)横放于其他竖直放置的图书上，读取这些图书的RFID标签并进行预处理，建立二次曲线模型；
步骤5-2，利用SVM分类算法、朴素贝叶斯分类算法等分类算法对步骤5-1建立的二次曲线模型进行训练，得到分类模型；
步骤5-3，得到分类模型后，读取待识别图书RFID标签，进行预处理，建立二次曲线模型，将二次曲线模型放入分类模型中，进行分类判断，识别图书的放置姿态为垂直于书架放置或平躺于书架放置。
其中，步骤6包括如下步骤：
步骤6-1，如果步骤5-3中判断图书为平躺于书架放置，则使用RFID设备竖向移动读取放置在图书内的RFID标签信号；
步骤6-2，利用RSSI图书排序算法最终确认图书的位置，判断步骤5结果是否有误。
其中，步骤6-2包括如下步骤：
步骤6-2-1，将有相同EPC号的三元组归为一类，进行数据预处理和二次曲线拟合；
步骤6-2-2，得出拟合后的曲线的波峰即最高点所在的时间T，将所有RFID标签的时间T进行排序，得出RFID标签对应的每本图书的位置顺序；
步骤6-2-3，如果RFID标签的时间按序排列，且被判断为平躺于书架放置的图书的所在的时间为排在最后，则表明步骤5判断正确，否则说明步骤5判断有误，图书是垂直于书架放置。
有益效果：本发明利用RFID标签信号的RSSI信号用于图书放置姿态的识别，准确判断出图书放置于书架上的状态，检测出平躺于书架其他图书上的图书，方便了图书管理人员纠正图书的错误放置姿态，让图书馆图书处于整洁有序的状态。
附图说明
下面结合附图和具体实施方式对本发明做更进一步的具体说明，本发明的上述和/或其他方面的优点将会变得更加清楚。
图1是天线与书架的位置关系俯视图。
图2是横放于书架的图书与垂直放置于书架的图书的主视图。
图3是RSSI值随时间变化图。
图4是实施例的流程图。
图5是图书排序的流程图。
具体实施方式
实施例1
RFID技术包括三个部分：天线，阅读器和RFID标签。天线通过发射出射频信号查询RFID标签，射频信号到达RFID标签后会被标签反射回来，阅读器通过接收到的反射信号识别RFID标签，RSSI信号就是反射信号的一种。本实施例利用标签反射的RSSI信号值实现图书馆图书放置姿态的识别。距离与RSSI信号值得关系如下：
d＝10^((abs(RSSI)-A)/(10×n))
其中，d为计算所得距离，RSSI为接收信号强度(负值)，A为发射端和接收端相隔1米时的信号强度，n为环境衰减因子。由上述公式可以转换得出：
RSSI＝-log(10×n×d+A)
上述公式表明，距离d越小，RSSI值越大，在本实施例中也就意味着天线与图书RFID标签的距离越近，获取到的RSSI值越大。
在对图书馆书架上的图书进行放置位置姿态的识别的过程中，应该让天线平行放置在距离图书不远的位置，否则RSSI值对于一本图书在不同时刻的大小不会有很大的差别，影响模型的建立。在本实施例中天线距离图书的距离设置为10cm。在扫描图书的过程中，匀速地将天线从书架的一侧缓慢移动至书架的另一侧，俯视图如图1所示。
在图书馆中，图书有时候会被读者错误地横放在其他正确垂直放置的图书上，导致了图书馆图书排放紊乱，图2显示了这种放置状态的现象。
通过理论分析及实验，RSSI信号值随着时间变化规律如图3。
通过图3可以得出以下结论：
当天线靠近RFID标签时，读取到的RFID标签的RSSI值呈递增趋势；当天线远离RFID标签时，RSSI值呈递减趋势，在RFID标签处读取的RSSI值最大；
本实施例提供了一种基于RFID技术RSSI信号值的图书放置姿态识别方法，其中数据采集的过程如下：
1、在每本图书中事先嵌入一个RFID标签，标签中写有包括图书id号的epc号，通过读取RFID标签的epc号，从其中提取出相应的图书id号，再从后台数据库服务器中查询相应的id号对应的图书信息。
2、将天线放置在距离书架约10cm的距离，以恒定的速度平行于书架，缓慢从书架的一端横向移动至另一端，采集标签的数据。
3、每次采集到的信息都包括三个数据：RFID标签的EPC号epc，标签反射的RSSI信号rssi值和读取标签信号的时间t，将此三个数据标记为一个三元组p＝{epc，rssi，t}，保存在后台数据库文件中。
在对图书放置姿态进行探测之前，要先确定一个分类模型，以便后续步骤根据此分类模型分析判断图书的放置姿态，具体过程如下：
1、随机选取部分图书，将这些图书横放于其他正常垂直于书架放置的图书上，记录下这些横放图书的信息。
2、按照上述数据采集的过程，采集图书的信息，包括横放图书信息和垂直放置的图书信息。
3、对上述采集的信息进行数据的处理，分别建立二次曲线模型，用朴素贝叶斯算法、SVM算法等，对二次曲线模型进行分类，构建一个分类模型。
数据采集完毕以后，要对数据进行预处理和建立模型，具体过程如下：
1、将采集到的三元组根据epc号进行分类，将持有相同epc号的三元组划分为一类P，表示一本图书在不同时刻读取到的RSSI信号值。
2、将每类P中的三元组根据时间排序，三元组时间t小的三元组p在前，时间t大的三元组p在后。
3、由于环境因素对标签反射的RSSI值会产生干扰，造成数据的不理想，需要对数据进行平滑处理，每个三元组中RSSI的值取其和相邻两个三元组RSSI值的平均值，以此减小数据的抖动现象。
4、获取每类P中的最大RSSI值以及对应的时间t，标记为maxindex，此时间表明天线最靠近RFID标签的时刻；
5、由于环境因素的影响，每本图书对应的rssi值曲线中，可能在某些时间出现的RSSI值的点偏离曲线较远，出现一些离散的点或者小的波，所以要先对这些稀疏的点进行处理。对于每类P，由maxindex向前后的三元组遍历，如果相邻的两个三元组p的时间差Δt大于一个阈值，就剔除在此三元组之外的三元组的数据，以此去除较离散的数据。此处阈值根据时间的大小设置，在此实施例中，时间的单位是秒，所以此阈值设置为0.1秒。
6、只需要得到最高点所在的波，对于其他较小的波，应该剔除。对每类P，由maxindex向前后的三元组遍历：向前遍历过程中，三元组中的rssi值应该越来越小，如果遍历到rssi值开始变大的三元组，就剔除此三元组之前的三元组数据；向后遍历过程中，三元组中的rssi值也应该越来越小，如果遍历到rssi值开始变大的三元组，剔除此三元组之后的三元组数据。
7、在对数据进行预处理以后，建立数据的二次曲线模型，得到RSSI值相对时间变化的曲线模型，此模型应该是开口向下的二次曲线，开口越大，表明读取到的对应的RFID标签的时间越长，读取次数越多，意味着此RFID标签处于横放的姿态的概率越大。如果二次曲线的二次项系数a的绝对值与上述训练过程得出的阈值相差在一个范围内，在此实施例中，此范围设置在0.005以内，具体范围视具体环境设置。
根据二次曲线模型和分类模型，判断出图书的放置姿态，需要最后通过图书排序算法确认图书的放置状态，具体过程如下：
1、将天线放置在距离书架约10cm的距离，在判断有横向摆放的图书的位置处，以恒定的速度平行于书架，缓慢从该层的书架底端移动至该层书架的上端，采集标签的数据。
2、数据分类、预处理以及建立二次曲线模型的过程同图书放置姿态识别的数据处理过程。
3、根据建立好的二次曲线模型，取其最高点所在的时间，对所有RFID标签对应的时间进行从小到大的排序，如果RFID标签有时间先后顺序的差别，且被判断横放的图书所在的时间最大，则判断此图书确实是被错误的横放在了其他图书上，而没有被误判。
本实施例的基于RFID标签RSSI值的图书馆图书放置姿态识别流程如图4所示。
1、将RFID标签嵌入图书中，标签epc号和图书id号绑定，并将图书信息和后台数据库服务器绑定
图书中的RFID标签epc号的格式设置为：
0BCXXXXXXXX000000000
0：表示标签的类型为图书RFID标签
B：书的借出状态
0：表示未借出
1：表示借出
2：表示不准外借
C：表示长度，目前图书的编码长度为8
XXXXX：表示图书ID号，由图书的条形码获得
00000：因RFID编码要求，不足双字节需要进行补零操作；
在获取标签的epc号之后，提取epc号第4位开始的8位数字，作为图书的id号，后台数据库中包含图书的id信息和图书的具体信息如书名、作者等。从读取到的标签的epc号中提取出图书的id号以后，可以根据此id号到后台数据库进行图书信息的查询。
2、用天线扫描图书，获取图书内标签的数据
图书馆书架上的图书应该整齐排放，否则会影响识别的效果。将所有的图书摆放 整齐以后，让天线平行放置于距离图书10cm左右的位置。天线不能离图书太远，否则在移动的过程中，一本书返回的RSSI信号值不会有太大的区别，因为在一定范围内天线相对于图书的距离没有发生大的变化。移动的速度也不应该过快，移动速度快，天线读取标签的次数会减少，本实施例将速度控制在0.1m/s左右。
数据的采集应该分为两次，一次是采集训练数据，得出分类模型和阈值，为后面的识别过程做准备，另一次就是采集要识别放置姿态的图书的数据，所以相应的数据处理和模型建立过程也要进行两次。
3、对采集的数据进行分类、预处理，剔除杂质数据，为构建理想的模型建立基础
采集数据的过程中，往往是多本图书交替反射信号，所以后台数据库记录的数据通常是多本图书的数据交错排列，需要将这些交错的数据分类，提取出每本图书对应的三元组集合，采用的分类方法就是将有相同epc号的数据划分为一类，然后按照时间的大小在划分好的三元组集合内进行数据的排序。由于环境因素对RFID标签的RSSI信号的干扰较大，加上书的厚薄程度不同，每本书采集到的RSSI信号数据不会很理想，有杂质数据，需要通过预处理剔除。理想的数据排列应该呈现近似二次曲线的形状，所以需要剔除比较离散的点和小的波峰。
4、对预处理好的数据拟合曲线，建立二次曲线模型，识别图书放置姿态
根据预处理的数据构建二次曲线模型，此模型应该是开口向下的二次曲线，开口越大，表明读取到的对应的RFID标签的时间越长，读取次数越多，意味着此RFID标签处于横放的姿态的概率越大。模型的构建分为两次，一次用于数据训练得出分类模型和分类的阈值，另一次就是图书放置姿态的识别，如果识别过程中得到的曲线的二次项系数与分类模型得出的阈值相差在一定小的范围内，就可以判断此图书是横放的状态。
5.利用图书排序算法最终判断图书的放置姿态
上述过程的判断结果可能会因为环境因素对RSSI信号的影响，有一定的误差，造成判断错误，为了最后确定图书确实是横放在其他竖直放置的图书上的，需要通过二次判断来做出最终的判别。在判断有横放图书的位置，进一步竖直扫描图书，采集数据，对图书进行排序。如果在此位置只有一本竖直放置的图书，那么最终的排序结果应该有且只有一个时间，或者几个相似的时间值，那么就证明判断错误，图书放置正 确。如果在此位置有横放在其他图书上的图书，那么最后的排序结果应该是有超过一本图书的时间，且时间间隔相差不小，被判断为横放的图书所在的读取时间在后。
本实施例的基于RFID标签RSSI值的图书馆图书排序的流程如图5所示。
1.扫描图书，获取数据
不同于图书放置姿态识别的数据采集过程，图书排序扫描图书的方向与图书姿态识别过程中的扫描图书方向垂直，为从一层书架的底端向上扫描到该层书架的上端为止。其余步骤都和图书放置姿态的识别过程一样。
2.对数据进行分类，画出RSSI值曲线图
获取的数据是多本书的数据交错在一起，需要将数据进行提取分类，将有相同epc号的数据归为一类，代表一本书的RSSI标签在不同时间返回的RSSI值。将分好类的数据分别按照时间顺序得到RSSI值的变化曲线图。
3.对数据进行预处理
由于环境因素的干扰，再加上书的厚薄程度不同，导致上述步骤得到的曲线图会有离散的点，或者会出现一些小的波峰，这些都是对后续步骤的干扰因子，需要将这些点进行剔除，只留下拥有最高值的一段连续曲线。
4.曲线拟合，图书排序
预处理好的曲线还不是能对其进行分析的最好模型，需要对曲线要进行二次曲线拟合，得到完整的随时间先递增，后递减的波段，取其最大值所在的时间为T。对得到的T1、T2、T3...Tn按照从小到大的顺序进行排序，这个顺序就是标签所在的图书的位置顺序。
本发明提供了基于RFID的RSSI信号值的图书放置姿态识别方法，具体实现该技术方案的方法和途径很多，以上所述仅是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。本实施例中未明确的各组成部分均可用现有技术加以实现。

Claims (7)

  1. 基于RFID的RSSI信号值的图书放置姿态识别方法，其特征在于，包括如下步骤：
    步骤1，使用RFID设备读取放置在图书内的RFID标签，RFID标签包括：图书RFID标签的EPC号epc，RFID标签RSSI值rssi，读取RFID标签的时刻t，将以上三个信息表示为一个三元组p＝{epc,rssi,t}，代表一本图书在t时刻读取到的放置在其内部的RFID标签的RSSI值；
    步骤2，将带有相同EPC号的三元组归为一类，作为一本图书在不同时刻获取到的内部RFID标签的RSSI信号值；
    步骤3，对步骤2获取的三元组进行预处理；
    步骤4，对步骤3中预处理好的数据建立二次曲线模型，得到图书RFID标签RSSI值随时间的变化规律；
    步骤5，利用步骤4建立的二次曲线模型进行训练，得出分类模型；采集待识别图书RFID标签，根据分类模型，判断待识别图书的放置姿态为垂直于书架放置或平躺于书架放置；
    步骤6，利用排序算法最终确认图书的位置，判定步骤5是否判断有误。
  2. 根据权利要求1所述的方法，其特征在于，步骤2包括：将步骤1得到的三元组根据EPC号进行分类，将持有相同EPC号的三元组划分为一类P，表示一本图书在不同时刻读取到的RSSI信号值。
  3. 根据权利要求2所述的方法，其特征在于，步骤3包括如下步骤：
    步骤3-1，对步骤2中每类P中的三元组根据时间排序，三元组时间t小的三元组在前，时间t大的三元组p在后；
    步骤3-2，对数据进行平滑处理；
    步骤3-3，获取每类P中的最大RSSI值以及对应的时间t，标记为maxindex；
    步骤3-4，对于每类P，由maxindex向前后的三元组遍历，如果相邻的两个三元组p的时间差Δt大于一个阈值，就剔除在此三元组之外的三元组的数据；
    步骤3-5，对于每类P，由maxindex向前后的三元组遍历：向前遍历过程中，如果遍历到rssi值开始变大的三元组，就剔除此三元组之前的三元组数据；向后遍历过程中，如果遍历到rssi值开始变大的三元组，剔除此三元组之后的三元组数据。
  4. 根据权利要求3所述的方法，其特征在于，步骤4包括：对步骤3中预处理好的数据进行二次曲线拟合，建立二次曲线模型，得到图书RFID标签RSSI值随时间的变化规律，二次曲线模型表示为如下形式：
    rssi＝a×t2+b×t+c，
    其中待求解的a，b，c分别表示二次曲线的系数。
  5. 根据权利要求4所述的方法，其特征在于，步骤5包括如下步骤：
    步骤5-1，随机取出一定数量的图书横放于其他竖直放置的图书上，读取这些图书的RFID标签并进行预处理，建立二次曲线模型；
    步骤5-2，利用分类算法对步骤5-1建立的二次曲线模型进行训练，得到分类模型；
    步骤5-3，读取待识别图书RFID标签，进行预处理，建立待识别图书的二次曲线模型，将待识别图书的二次曲线模型放入分类模型中，进行分类判断，识别图书的放置姿态为垂直于书架放置或平躺于书架放置。
  6. 根据权利要求5所述的方法，其特征在于，步骤6包括如下步骤：
    步骤6-1，如果步骤5-3中判断图书为平躺于书架放置，则使用RFID设备竖向移动读取放置在图书内的RFID标签；
    步骤6-2，利用RSSI图书排序算法最终确认图书的位置，判断步骤5结果是否有误。
  7. 根据权利要求6所述的方法，其特征在于，步骤6-2包括如下步骤：
    步骤6-2-1，将有相同EPC号的三元组归为一类，进行数据预处理和二次曲线拟合；
    步骤6-2-2，得出拟合后的曲线的波峰即最高点所在的时间T，将所有RFID标签的时间T进行排序，得出RFID标签对应的每本图书的位置顺序；
    步骤6-2-3，如果RFID标签的时间按序排列，且被判断为平躺于书架放置的图书的所在的时间为排在最后，则表明步骤5判断正确，否则说明步骤5判断有误，图书是垂直于书架放置。
PCT/CN2017/070930 2016-05-13 2017-01-11 基于rfid的rssi信号值的图书放置姿态识别方法 Ceased WO2017193624A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/739,014 US10282576B2 (en) 2016-05-13 2017-01-11 Book placing attitude recognition method based on RSSI signal value of RFID

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610318946.1A CN106022404B (zh) 2016-05-13 2016-05-13 基于rfid的rssi信号值的图书放置姿态识别方法
CN201610318946.1 2016-05-13

Publications (1)

Publication Number Publication Date
WO2017193624A1 true WO2017193624A1 (zh) 2017-11-16

Family

ID=57100455

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/070930 Ceased WO2017193624A1 (zh) 2016-05-13 2017-01-11 基于rfid的rssi信号值的图书放置姿态识别方法

Country Status (3)

Country Link
US (1) US10282576B2 (zh)
CN (1) CN106022404B (zh)
WO (1) WO2017193624A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114707620A (zh) * 2022-04-15 2022-07-05 天津工业大学 一种基于相位修正的超高频rfid相对定位方法
CN117315864A (zh) * 2023-11-29 2023-12-29 青岛中科力天网络科技有限公司 一种ai智慧图书馆管理系统

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106022404B (zh) * 2016-05-13 2018-04-03 南京大学 基于rfid的rssi信号值的图书放置姿态识别方法
CN109190426A (zh) * 2018-07-30 2019-01-11 太原理工大学 一种基于rfid的图书自动检测和定位方法
US11611584B2 (en) 2019-03-01 2023-03-21 Cloudentity, Inc. Smart bits
CN110008766B (zh) * 2019-03-26 2020-04-24 南京大学 一种基于射频识别技术的室内人数统计方法
CN110348258B (zh) * 2019-07-12 2020-11-24 西安电子科技大学 一种基于机器学习的rfid应答信号帧同步系统及方法
CN110542882A (zh) * 2019-08-27 2019-12-06 南京理工大学 基于rfid的抽屉内物品位置计算方法
CN112183586B (zh) * 2020-09-10 2024-04-02 浙江工业大学 在线多任务学习的人体姿态射频识别方法
CN112200282A (zh) * 2020-10-14 2021-01-08 上海海事大学 基于特征加权支持向量机的rfid智能图书定位方法
CN112464682B (zh) * 2020-11-26 2022-05-10 上海阿法迪智能数字科技股份有限公司 基于rfid技术的智能书架的图书定位装置、方法
CN112560514A (zh) * 2020-12-15 2021-03-26 天津光电通信技术有限公司 一种基于rfid的倾斜物体检测方法
CN113639816A (zh) * 2021-07-06 2021-11-12 燕山大学 非侵入式的容器内液位高度监测方法及实现所述方法的系统
CN113962317B (zh) * 2021-10-28 2026-04-28 阿依瓦（北京）技术有限公司 图书定位方法
CN117422079B (zh) * 2023-10-18 2024-08-02 上海阿法迪智能数字科技股份有限公司 基于盘点机器人的图书盘点方法和系统
CN121580144B (zh) * 2026-01-26 2026-04-10 南京信息工程大学 一种基于多层级标签协同感知的床上人体姿态识别方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140266608A1 (en) * 2013-03-14 2014-09-18 Wal-Mart Stores, Inc. Method and apparatus pertaining to the identification of physical-local discrepancies in statistics-based rfid-tag information
CN104915618A (zh) * 2015-07-03 2015-09-16 南京大学 一种基于超高频rfid技术的图书馆典藏方法
CN104933458A (zh) * 2015-07-07 2015-09-23 南京大学 一种超高频rfid推车式图书智能上架盘点设备
CN106022404A (zh) * 2016-05-13 2016-10-12 南京大学 基于rfid的rssi信号值的图书放置姿态识别方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6655586B1 (en) * 2000-02-25 2003-12-02 Xerox Corporation Systems and methods that detect a page identification using embedded identification tags
US7621448B2 (en) * 2004-09-30 2009-11-24 Ricoh Company, Ltd. Information display medium, information managing apparatus, information managing method, guidance managing method, and guidance managing program
CA2875578A1 (en) * 2014-12-24 2016-06-24 Stephan HEATH Systems, computer media, and methods for using electromagnetic frequency (emf) identification (id) devices for monitoring, collection, analysis, use and tracking of personal, medical, transaction, and location data for one or more individuals
US20160189170A1 (en) * 2014-12-29 2016-06-30 International Business Machines Corporation Recognizing Customers Requiring Assistance

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140266608A1 (en) * 2013-03-14 2014-09-18 Wal-Mart Stores, Inc. Method and apparatus pertaining to the identification of physical-local discrepancies in statistics-based rfid-tag information
CN104915618A (zh) * 2015-07-03 2015-09-16 南京大学 一种基于超高频rfid技术的图书馆典藏方法
CN104933458A (zh) * 2015-07-07 2015-09-23 南京大学 一种超高频rfid推车式图书智能上架盘点设备
CN106022404A (zh) * 2016-05-13 2016-10-12 南京大学 基于rfid的rssi信号值的图书放置姿态识别方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, XIANGCHEN: "A book auto inventory method design and implementation based on variation of RSSI in RFID library", ELECTRONIC TECHNOLOGY & INFORMATION SCIENCE, CHINA MASTER'S THESES FULL-TEXT DATABASE, no. 3, 15 March 2015 (2015-03-15), pages 28 - 41 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114707620A (zh) * 2022-04-15 2022-07-05 天津工业大学 一种基于相位修正的超高频rfid相对定位方法
CN117315864A (zh) * 2023-11-29 2023-12-29 青岛中科力天网络科技有限公司 一种ai智慧图书馆管理系统
CN117315864B (zh) * 2023-11-29 2024-03-08 青岛中科力天网络科技有限公司 一种ai智慧图书馆管理系统

Also Published As

Publication number Publication date
US20180189529A1 (en) 2018-07-05
CN106022404B (zh) 2018-04-03
US10282576B2 (en) 2019-05-07
CN106022404A (zh) 2016-10-12

Similar Documents

Publication Publication Date Title
WO2017193624A1 (zh) 基于rfid的rssi信号值的图书放置姿态识别方法
JP4883649B2 (ja) 画像認識方法、画像認識装置および画像認識プログラム
CN104915618B (zh) 一种基于超高频rfid技术的图书馆典藏方法
US9443164B2 (en) System and method for product identification
CN105528604B (zh) 一种基于ocr的票据自动识别与处理系统
US7356162B2 (en) Method for sorting postal items in a plurality of sorting passes
EP3008666B1 (en) Image based object classification
JP6693059B2 (ja) 商品棚認識装置、商品棚認識方法、プログラム及び画像処理装置
US10275620B2 (en) Book sorting method on the basis of RSSI signal value of RFID tag
KR101878239B1 (ko) 모바일 로봇 기반의 도서 관리 시스템
CN109376820A (zh) 基于rfid技术的图书精确定位排序方法
Namane et al. Fast real time 1D barcode detection from webcam images using the bars detection method
CN105701518A (zh) 基于rfid工器具智能管理装置
Narwade et al. Offline handwritten signature verification using cylindrical shape context
CN113269126A (zh) 一种基于坐标转换的关键信息提取方法
US8184895B2 (en) Method and apparatus for magnetic character recognition
Duan et al. Identifying books in library using line segment detector and contour clustering
CN108557364B (zh) 一种自动盘库方法和装置
Nguyen et al. Symbol descriptor based on shape context and vector model of information retrieval
CN112703508B (zh) 用于辨识物体的方法和设备
CN114604543B (zh) 一种计量识别系统、计量识别方法及周转箱
Chatbri et al. Shape matching using keypoints extracted from both the foreground and the background of binary images
CN109359705A (zh) 一种错位书籍查找系统及其查找方法
CN115169375A (zh) 基于ar与枪球联动的高位物料可视化方法
AhilaPriyadharshini et al. Object recognition based on local steering kernel and SVM

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 17795251

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17795251

Country of ref document: EP

Kind code of ref document: A1