TW201101241A - A falling down detection method implemented by tri-axis accelerometer - Google Patents

A falling down detection method implemented by tri-axis accelerometer Download PDF

Info

Publication number
TW201101241A
TW201101241A TW98121994A TW98121994A TW201101241A TW 201101241 A TW201101241 A TW 201101241A TW 98121994 A TW98121994 A TW 98121994A TW 98121994 A TW98121994 A TW 98121994A TW 201101241 A TW201101241 A TW 201101241A
Authority
TW
Taiwan
Prior art keywords
axis
fall
acceleration
ward
state
Prior art date
Application number
TW98121994A
Other languages
Chinese (zh)
Other versions
TWI374408B (en
Inventor
Hsu-Yang Kung
Chin-Yu Ou
Che-I Wu
Hung-Chin Wang
Chun-Hao Lin
Shin-Di Li
Miao-Han Chang
Yu-Lun Syu
Hong-Jie Chen
Original Assignee
Univ Nat Pingtung Sci & Tech
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 Univ Nat Pingtung Sci & Tech filed Critical Univ Nat Pingtung Sci & Tech
Priority to TW98121994A priority Critical patent/TWI374408B/en
Publication of TW201101241A publication Critical patent/TW201101241A/en
Application granted granted Critical
Publication of TWI374408B publication Critical patent/TWI374408B/en

Links

Landscapes

  • Emergency Alarm Devices (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

A falling down detection method implemented by tri-axis accelerometer is presented. The detection method comprises a state information collecting step, a falling down condition determining step and an emergency announcing step. The state information collecting step collects the state information generated from the change in posture of a ward by using at least one tri-axis accelerometer. The falling down condition determining step determines whether any value in the state information goes beyond a predetermined normal range by a monitor unit to judge whether the ward falls down or not and if the ward can self-restore posture or not. Besides, once a falling down of the ward is detected, a timer is started. The emergency announcing step decides to pass information regarding to the falling down of the ward or not according to the result of the timer.

Description

201101241 六、發明說明: 【發明所屬之技術領域】 本發明係關於一種應用三軸加速規之跌 法,尤其是一種藉由裝設於人體之三軸加迷賴^偵剎方 倒偵測方法。 ^丁的跌 【先前技術】 根據國民健康保險局的統計,意外傷害 0丨☆题死亡率 的比率已逐年增加;其巾無論男性或女性老年人因 外事故所造成的生命損害,僅次於交通意外事故,排在二 外傷害弟一位。此外’由於全球性人口的老化問題 照濩已成為未來醫療產業中必然發展的重要領域.因此 如何加速老人照護醫療科技的發展,已成為現代相當 的醫療防護議題之一。 w 要 更詳言之,老人跌倒比率佔總人口跌酬造成的傷 害比率近百分之五十,嚴重者甚至會導致死亡,亦成為跌 倒傷害的最大宗事件。跌倒意外事故會造成如&重大的傷 害’往往取決於緊急醫療的時間;因此,偏若可有效债測 跌倒意外事故的發生,方可掌握送醫的「黃金時間」,'二 降低傷害比率。 」 、舉例而言,目前常見之針對跌倒意外事故的習知偵 測方法大致包含有以下數種: 第一種習知偵測方法,如中華民國公開第2〇〇912814 號配戴式活動感測裝置及其方法」,該感測裝置包含一 腕錶及一感測元件,該腕錶及感測元件均設有三轴加速度 201101241 ㈣量測身軀姿勢變換與活動加速度值,且其姿 如則判斷為跌倒’可發出警報及通知緊急救護, 如為誤判時,必触配戴者自行取消警報。 2 ’上述第-種習知仙方以須㈣腕錶及感 跌:m人體正雜置後,方可配戴者是否 ,者,錢行該侧方法時,倘賴腕錶及該感測 =之女裝位置不正確,將無法進行偵測;因此,具有安 便之缺點。另外,該習㈣測方法係以量測身躯 k換與_加速度值,關斷配戴者是否跌倒,由於 且'配合參照如錄改變之加速㈣鱗目素,故相當容 易產生誤判情形。201101241 VI. Description of the Invention: [Technical Field] The present invention relates to a method for applying a three-axis acceleration gauge, and more particularly to a method for detecting a tilt by means of a three-axis plus fascination. ^Ding's fall [previous technology] According to the National Health Insurance Bureau, the rate of death from accidental injury 0丨☆ has increased year by year; its towel is second only to the life damage caused by accidents caused by male or female elderly people. Traffic accidents, ranked second in the injury to a younger brother. In addition, due to the aging of the global population, care has become an important area of future development in the medical industry. Therefore, how to accelerate the development of medical care technology for the elderly has become one of the most modern medical protection issues. w To be more specific, the rate of injury caused by the fall rate of the elderly is nearly 50%, which is the worst cause of falls. Falling accidents can cause serious injuries such as & often depending on the time of emergency medical treatment; therefore, if you can effectively measure the occurrence of a fall accident, you can grasp the "golden time" of sending the doctor, and the second is to reduce the injury rate. . For example, the conventional detection methods for fall accidents generally include the following: The first conventional detection method, such as the Republic of China Open No. 2, No. 912814, wearing a sense of activity Measuring device and method thereof, the sensing device comprises a wristwatch and a sensing component, wherein the wristwatch and the sensing component are respectively provided with a three-axis acceleration 201101241 (four) measuring body posture change and active acceleration value, and the posture is as follows If it is judged to be a fall, it can issue an alarm and notify the emergency ambulance. If it is a misjudgment, the wearer must cancel the alarm. 2 'The above-mentioned first-class knowledge of the immortal square to the (four) watch and the feeling of falling: m body is mixed, can the wearer be the person, if the money is on the side of the method, if the watch and the sensor = If the women's position is incorrect, it will not be detected; therefore, it has the disadvantage of being comfortable. In addition, the method of measuring (4) is to measure the body k and the _acceleration value to turn off whether the wearer falls, and because of the acceleration of the reference (4) squama, it is quite easy to produce a false positive situation.

G 第二種習知债測方法,如中華民國公開第观期63 ^可攜式行為監_統及其方法」,係湘具有三轴加 j的可攜式資料擷取單元收較财之行為監师料, ,错由無線傳輸將行為監測資料傳送至分散式資料祠服 盗’然後由分散式資料舰n分析行為監測資料, ^測者之動作事件。若_出受測者發生異常之動作事 件,再傳送異常事件訊息至電子接收裝置。 ^而’上述第二種習知_方法係將該可攜式資料 擷取I元絲於人體’域測_者身料勢及姿勢變化 :斤產,的加速度變化,以便輸出行為監測資料,再利用分 政式資料彳嫌ϋ分析行為監崎料,用關斷配戴者是否 ,倒;惟該習知_方法僅單純針對配戴者身體姿勢變 用以判斷配戴者是否發生異常之動作,因此,亦容易 201101241 M337!^知偵測方法,係藉由如中華民國公告第 細m目r—人體.平衡位移量測系統」,該系統包含一三 r 口盆“I減法器、-放大電路一資料#貞取卡及-電 ::乂稭由該系統所執行之習知偵測方法,當該三軸加 Γ=:夺間内移動時,可產生-訊號以記錄加速度資 取卡轉換器=放大器處理後傳送至該資料擷 、田°式最後由该電腦加以計算並緣製成圖 形,以ί、判讀人體平衡狀態。 、= ί述第三㈣知_方法在#由三轴加速規 ^录加速度t料後,仍必須透過該減法器、該放大電路、 =料擷取卡及該電腦的娜運算處理,方可觸人體平 ’故具有步驟流程複雜等缺點,且同樣容易產生誤 味「f四Γ習知摘測方法,如中華民國公告帛腿779 唬七、配帶於腰部的健康管理及跌倒警示裂置」,該警示 A,包括處理#組、監測模組、傳輸彳脸及警示模組。並 中藉由該警示U所執行㈣知偵财法,侧用監測模 組將使用者之独狀況訊息雜錢理顧,當使用者跌 倒時,處理模組會接收到使用者跌倒訊息,並藉由警示模 組提出警不’或者經由傳輸模組發出簡訊通知救助單元。 然而,上述第四種習知_方法必_警示裝置安 裝於使用者腰部位置,方可降低制跌峨態之誤差,因 此,在安纽用上受_t限制,造成許&便。另外, 該監測模組储由侧使用者之運紐況,以判斷使用者 是否有跌倒情形,而同樣容易產生誤判情形。 6 201101241 第五種習知偵測方法,如 號「蠖”涵郝、家缺„ 中華民國公開第200910268 :路=」,該遙控器包括加速感測器、觸動 ί 二動介面電路、处讎器及微控 控器執所執行的習知偵測方法,係利用微 3:人d播#、汁時及輪入輸出埠控制等步驟,且 以達到計步、緊急呼救、跌土 區疋位等功能。 ΟG The second method of traditional debt measurement, such as the Republic of China public open period 63 ^ portable behavior supervision _ system and its methods, the Department of Hunan has a three-axis plus j portable data acquisition unit to collect money The behavior supervisor, the error, is transmitted by wireless transmission to the decentralized data, and then the behavioral monitoring data is analyzed by the distributed data ship n. If an abnormal event occurs in the test subject, an abnormal event message is transmitted to the electronic receiving device. ^ And 'the above-mentioned second conventional method _ method is to take the portable data to the I body in the human body's body measurement _ body body posture and posture changes: the production of acceleration, change in order to output behavior monitoring data, Re-use of the sub-political information, suspicion of the analysis of the behavior of the monitoring of the raw materials, to shut down whether the wearer is down, but the conventional method _ method is only for the wearer's body posture to determine whether the wearer is abnormal Action, therefore, is also easy 201101241 M337! ^ know detection method, such as the Republic of China Announcement of the fine m-r-body. Balanced displacement measurement system, the system contains a three-r pot "I subtractor, -Amplification circuit-data#Capture card and -Electricity:: The traditional detection method performed by the system, when the three-axis twisting =: movement within the zone, a signal can be generated to record the acceleration The card-receiving converter = the amplifier is processed and transmitted to the data, and the field is finally calculated by the computer and patterned into a figure to interpret the balance state of the human body. == ίThe third (four) knowledge_method in #由After the three-axis acceleration regulation is recorded, the subtractor must still pass through the subtractor. The amplifying circuit, the material picking card and the computer's nano processing can only touch the human body, so it has the disadvantages of complicated step processes, and it is also easy to produce misleading "f Γ Γ 摘 摘 , ,, such as the Republic of China announcement帛 779 、 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , In the meantime, the warning module U performs the (4) knowledge-detection method, and the side-use monitoring module ignores the user's unique status message. When the user falls, the processing module receives the user's fall message, and The warning module is used to alert the rescue unit by sending a warning message via the transmission module. However, the above-mentioned fourth conventional method _method must be installed at the waist position of the user to reduce the error of the squatting state, and therefore, it is limited by _t in the use of Annu, causing the & In addition, the monitoring module stores the state of the user by the side to determine whether the user has a fall situation, and is also prone to misjudgment. 6 201101241 The fifth method of detecting the familiar, such as the number "蠖" Han Hao, the home is missing „Republic of China Open 200910268: Road=”, the remote control includes acceleration sensor, touch ί two-touch interface circuit, The conventional detection method performed by the device and the micro controller is based on the steps of micro 3: human d broadcast #, juice time and wheel input and output control, and to achieve step counting, emergency call, and falling soil area. Bit and other functions. Ο

fTO:、而i述第五種習知偵測方法主要制用加速感 叙=測人體所產生的加速度變化,且配合職電路的觸 〜^號產生電路啟動’再依程式轉分析關定使用者為 二,或跌倒’並透過RP收發信器發出跌倒通報;因此, 4知制方法僅為單純考慮人體所錢的加速度變,化, 用以=斷使用者是否有跌倒情形,故料產生誤判情形。 > 4上所述,上述各種可針對跌倒意外事故進行偵測 的=知_方法,大致具有絲使料便、躺品質不佳 及谷易產生誤判等料缺點,故仍有改良之必要。 【發明内容】 本發明目的乃改良上述缺點,以提供一種應用三軸 加速規之跌倒债測方法’可藉由被監護者因身體姿勢改變 所產生的、、加速度變化〃及、、角度變化〃,以達到降低跌 匈偵測誤判情形之目的。 本發明次一目的係提供一種應用三軸加速規之跌倒 谓測方法’可利用計時手段進一步確認先前之跌倒情形是 否有誤判情形,以達到提升偵測品質之目的。 201101241 本發明另一目的係提供一種應用三轴加速規之跌倒 偵測方法,可藉由主動式通報機制,以達到降低意外傷害 比率之目的。 根據本發明應用三轴加速規之跌倒偵測方法,包含 下列步驟: 一狀態資訊擷取步驟,係藉由至少一個三軸加速規 搜集被監護者因身體姿勢改變所產生的加速度變化及角度 變化’以獲得被監護者之狀態資訊; 一跌倒條件判斷步驟’係藉由一分析監控單元判斷 δ亥狀,%資訊之加速度變化是否超過一跌倒判定門檻值,該 跌倒判定門檻值為3.55g到3.62g之間,且利用該狀態資 訊之角度變化判斷被監護者是否能自行站立,並在判斷無 法自行站立後進入一計時狀態,而判斷能自行站立後則重 新回到該狀態資訊擷取步驟;及 一緊急通報步驟,係依據該計時狀態之結果決定是 否回傳被監護者的跌倒資訊。 【實施方式】 為讓本發明之上述及其他目的、特徵及優點能更明 ,易懂,下文特舉本發明之健實齡!,並配合所附圖 式,作詳細說明如下: 請參圖所示’本發明應用三轴加速規 _方法健可雜監護者身體之輕雜錢至少—個 三軸加速規11,用简時㈣被 所產生的 '、狀態資訊八詳袁 勢改邊 ^俊π 5兒明),並藉由一分析監 201101241 控單元12針對、、狀態資訊〃進行分析運算處理,以判斷 被監s蔓者是否有跌倒情形。其中: Ο Ο 該三軸加速規11之詳細結構、運作方式及所能達成 的功冑b係屬習知技術’容不贅述。再者,市面上常見之三 軸加速規11大致可分為兩種,第一種三軸加速規u在靜 止狀態下可受到地力引力lg (加速度)的作用力,第二 種三軸加速規11在靜止狀態下不會受到地力引力lg (加 速度)的作用力;在如圖所示之實施例中,本發明係採用 第一種三軸加速規11,合先敘明。另外,請配合參照第2 圖所示,本發明所採用之三軸加速規11具備有X軸、γ 轴及Z軸,其中同一軸的兩個方向分別以+和—符號予 以區別,例如:該三轴加速規U提供有χ+、x—、γ +、Υ—、Ζ+、Ζ—等六個感測方向。 舉例而§,當X軸與Υ軸平行於地面(即X軸與Υ 軸輸出為0g),且未受到任何外力作用時,則ζ+或L 會收到lg的地心引力作用;反之料χ軸與ζ軸平行於 地面,則Υ軸之Υ+、γ—感測方向會收到ig的地心引 力^此,以便利用X幸由、Y軸及z軸所產生的變化, 予^搜集被監護者因身體姿勢改變所產生的、、狀態資 =一如第2 @所TF ’係揭示當被監護者有跌倒情形時, u三軸加速規n 轴、γ軸及z軸則會產生變化。 該分析監控單元12用以分析該三轴加速規u所搜 狀態資訊〃’以便_被監護者是否有跌倒情形, ^即時將各雛料魏者;科,該分析監控單 較佳可架構於各種行動式平台,例如:智慧型手機 201101241 戈仃動數位助理(PDA)等,以方便政 即時察看被監護者的壯能. m 蔓者可 ㈣的狀癌,或者,亦可依監護者的實際兩 求,將该分析監控單元u架構於後端飼服器。貫不而 叫參照第1及3圖所示’本發明應用三軸加速規之 跌倒偵測方法储由該三軸加速規u及分析監元 執行以下步驟’其步驟至少包含—狀態資訊擷取步驟 、-跌倒條件判斷步驟S2及一緊急通報步驟幻。 、該狀態資_取步驟S1主要係該分析監控單元 12透過該三軸加速規n _搜#被監護相身體姿 變所產生的、'加速度變化"及、、角度變化", 護者之、、狀態資訊〃。如第4圖所示,係設贿性座標之 :軸為X,.軸、γ,軸及z,軸,該慣性座標係指地心引力的 二轴方向;人體座標之三軸為X軸、γ軸及z軸,該人 體座標係指該三軸加速規U的三軸方向(隨被監護者因 身體姿勢改變而產生變化)。因此,、、加速度變化及、、角 度變化〃可健者因身體雜改,該人體座標相 對於慣性座標所產生的變化所獲得。 更詳言之,該狀態資Wi頁取步驟S1較佳可區分如第 5圖所示之步驟S11及步驟S12,其中: 步驟S11係接收該三軸加速規丨丨之三軸軸、γ 轴及ζ軸)的加速度分量並轉換為加速度(3),以便後 續執行步驟S12。 步驟S12係計算人體座標之三軸(χ軸、γ軸及ζ 轴)與慣性座標之三軸(X,軸、Υ,軸及ζ,軸)彼此之間 的角度;舉例而言,X軸與X,軸的夾角為α,γ軸與γ, —10 — 201101241 轴的夾角為β,ζ轴與ζ’軸的夾角為r (如第4圖所 示),以便後續執行該跌倒條件判斷步驟S2。 藉由上述步驟S11及步驟S12,即可分別獲得、、加速 度變化〃及 '、角度變化〃;由於上述該三輛加速規η所接 收的加速度(2)為三軸(X軸、γ軸及ζ軸)之各別加 速度分量();因此,如步驟sii,在計算夾角 α、夾角召及夾角7前’必需事先透過歐幾里德距離公式 (如公式1所示)計算加速度(3),其中(χ,Ρ,芝)係 ΟfTO:, and the fifth conventional detection method mainly uses the acceleration sense=measures the acceleration change generated by the human body, and cooperates with the touch circuit of the occupation circuit to generate the circuit start'. The second is, or falls, and sends a fall notification through the RP transceiver; therefore, the 4 knowledge method is simply to consider the acceleration of the human body's money, and to use = to break the user whether there is a fall situation, so the material is generated Misjudgment. > As described above, the above-mentioned various methods for detecting a fall accident have a shortcoming such as shortness of the material, poor quality of the lying, and misjudgment of the valley, so that there is still a need for improvement. SUMMARY OF THE INVENTION The object of the present invention is to improve the above disadvantages, and to provide a method for measuring a fall of a debt using a three-axis accelerometer, which can be caused by a change in body posture due to a change in body posture, an increase in acceleration, and an angle change. In order to achieve the purpose of reducing the false positives of the Hungarian detection. The second object of the present invention is to provide a fall prediction method using a three-axis acceleration gauge, which can be used to further confirm whether a previous fall situation is misjudged to achieve the purpose of improving detection quality. 201101241 Another object of the present invention is to provide a fall detection method using a three-axis acceleration gauge, which can achieve the purpose of reducing the accidental injury rate by an active notification mechanism. According to the present invention, a fall detection method for applying a three-axis accelerometer includes the following steps: A state information acquisition step of collecting acceleration changes and angle changes caused by a change in body posture of a ward by at least one three-axis acceleration gauge Obtaining status information of the ward; a fall condition determination step 'determines whether the acceleration change of the % information exceeds a fall determination threshold by an analysis monitoring unit, and the fall determination threshold is 3.55 g to 3.62 Between g, and using the change of the angle of the state information to determine whether the ward can stand on its own, and after entering the state of judging that it is unable to stand on its own, and judging that it can stand on its own, then returning to the state information acquisition step; And an emergency notification step is based on the result of the timing state to determine whether to return the ward's fall information. [Embodiment] The above and other objects, features and advantages of the present invention will become more apparent and understood. And with the accompanying drawings, a detailed description is as follows: Please refer to the figure below. 'The invention applies the three-axis accelerometer _ method, the health and the miscellaneous guardian's body light money at least one three-axis acceleration gauge 11, with Jane (4) The resulting ', the state information is detailed, the Yuan potential changes the edge ^ Jun π 5 children Ming), and through an analysis of the 201101241 control unit 12, the state information 〃 analysis and processing, to judge the s Whether there is a fall situation. Among them: Ο 详细 The detailed structure, operation mode and the achievable work of the three-axis accelerometer 11 are all well-known technologies. Furthermore, the three-axis accelerometer 11 that is common in the market can be roughly divided into two types. The first three-axis accelerometer u can be subjected to the force of gravitational force lg (acceleration) in a static state, and the second triaxial accelerometer 11 is not subjected to the force of gravitational force lg (acceleration) in the stationary state; in the embodiment as shown, the first triaxial accelerometer 11 is used in the present invention. In addition, as shown in Fig. 2, the three-axis acceleration gauge 11 used in the present invention has an X-axis, a γ-axis, and a Z-axis, wherein the two directions of the same axis are distinguished by the + and - symbols, for example: The three-axis acceleration gauge U provides six sensing directions of χ+, x-, γ+, Υ-, Ζ+, Ζ-. For example and §, when the X axis and the Υ axis are parallel to the ground (ie, the X axis and the Υ axis output are 0g), and no external force is applied, ζ+ or L will receive the gravity of lg; The χ axis and the ζ axis are parallel to the ground, then the Υ+, γ-sensing direction of the Υ axis will receive the gravity of ig ^, in order to utilize the changes produced by X, Y and z axes, to ^ Collecting the ward's body due to changes in body posture, the status of the state = as the second @ TF ' reveals that when the ward has a fall, u triaxial acceleration gauge n axis, γ axis and z axis will Make a difference. The analysis monitoring unit 12 is configured to analyze the status information of the three-axis acceleration gauge 〃' so that the Guardian has a fall situation, and the individual will be ready for the child; Various mobile platforms, such as: smart phone 201101241 Ge 仃 digital assistant (PDA), etc., in order to facilitate the government to see the executor's strength. m vine can (4) cancer, or, depending on the guardian Actually, the analysis monitoring unit u is constructed on the back end feeder. Referring to Figures 1 and 3, the fall detection method for applying the triaxial acceleration gauge of the present invention stores the following steps by the three-axis acceleration gauge u and the analysis supervisor. The steps include at least the status information acquisition step. - The fall condition determination step S2 and an emergency notification step are magical. The state _ taking step S1 is mainly the 'acceleration change' and the angle change " generated by the analysis and monitoring unit 12 through the three-axis acceleration gauge n _ search# , status information 〃. As shown in Figure 4, the british coordinates are: the axis is X, the axis, the γ, the axis and the z, the axis, the inertial coordinate is the two-axis direction of the gravity; the three axes of the human coordinate are the X-axis The γ-axis and the z-axis, the human body coordinate refers to the three-axis direction of the three-axis acceleration gauge U (as the ward is changed due to changes in body posture). Therefore, the change in acceleration, and the change in angle can be obtained by the change of the body coordinates relative to the inertial coordinates due to physical changes. More specifically, the state Wi-page taking step S1 preferably distinguishes between step S11 and step S12 as shown in FIG. 5, wherein: step S11 receives the three-axis axis and the γ-axis of the three-axis acceleration rule. The acceleration component of the ζ axis is converted into acceleration (3) for subsequent execution of step S12. Step S12 is to calculate the angle between the three axes of the human body coordinate (the χ axis, the γ axis, and the ζ axis) and the three axes of the inertial coordinate (X, axis, Υ, axis, and ζ, axis); for example, the X axis The angle between the axis and the axis is α, the axis of γ is γ, the angle between the axis of 1010 and 201101241 is β, and the angle between the axis of ζ and ζ' is r (as shown in Fig. 4), so that the fall condition can be judged subsequently. Step S2. By the above steps S11 and S12, the acceleration changes 〃 and ', the angle change 分别 can be respectively obtained; since the acceleration (2) received by the three acceleration gauges η is three axes (X axis, γ axis and The respective acceleration components of the ζ axis); therefore, as in step sii, before calculating the angle α, the angle of the angle and the angle of the angle 7, the acceleration must be calculated in advance through the Euclidean distance formula (as shown in Equation 1) (3) , where (χ,Ρ,芝) system

代表三軸加速規U之三軸(χ軸、γ軸及ζ軸)所接收Represented by the three axes of the three-axis acceleration gauge U (χ axis, γ axis and ζ axis)

的加速度分量(g值)。另外,如步驟S12,再計算X 軸、Y轴及Z軸(人體座標)相對於χ’軸、γ,軸及z,轴 (慣性座標)的夾角α、夾角万及夾角r (如公式2所 不)’爽角α、失角/5及夾角r的定義方式可依據表1之 規格自°丁。藉由失角α、夾角/5及爽角r的定義之後,即 可在不考量被監護者配戴該三軸加速規11之角度的情況 下找出地引力方向,以方便被監護者安裝該三軸加速 規11。 λ/^2 + 72 + Ζ2 ……(公式 1) γ ΖThe acceleration component (g value). In addition, as in step S12, the X angle, the Y axis, and the Z axis (human coordinates) are calculated relative to the χ' axis, γ, the axis, and the axis (inertial coordinate), the angle α, the angle angstrom, and the angle r (as in Equation 2). No) The definition of 'slow angle α, lost angle /5 and angle r can be determined according to the specifications of Table 1. By defining the angle of absence α, the angle /5 and the refresh angle r, the direction of gravity can be found without considering the angle of the three-axis accelerometer 11 to facilitate the installation of the guardian. The three-axis acceleration gauge 11. λ/^2 + 72 + Ζ2 ......(Formula 1) γ Ζ

~J,fi = COSSy = COSS......(公式 2) a A cos~l (表1) 箭頭軸向 反向地面f 平行地面-^ ----— 正向地面j 1------ a ~\ a =90。 »___ 〇: =180° x+ Y+ Z + β=〇。 β =90' = 180' 7=0。 r =90。 T —180' —11 — 201101241 該跌倒條件判斷步驟S2主要係判斷加速度(2)是 否超過、、跌倒判定門檻值〃,藉以判定被監護者是否發生 跌倒情形,並利用加速度(2)的變化,進一步區分成跌 倒狀態(Falldown)與正常狀態(Normal)。由於二倒時 間往往在一秒鐘内就倒地,因此,本發明除利用加速^ (3)予以判斷被監護者之跌倒力道外,更進—步 ^ 續多次的加速度(2)取樣來進行判斷。如第6 ^ 曲線(L)為跌倒前後之加速度(2) 以#却·中μ、+>、、 日7遷、戈變化,藉此~J,fi = COSSy = COSS......(Formula 2) a A cos~l (Table 1) Arrow Axis Reverse Ground f Parallel Ground -^ ----- Forward Ground j 1-- ---- a ~\ a =90. »___ 〇: =180° x+ Y+ Z + β=〇. β = 90' = 180' 7 = 0. r = 90. T —180′ —11 — 201101241 The fall condition determination step S2 mainly determines whether the acceleration (2) exceeds or falls the determination threshold value 〃, thereby determining whether the ward has a fall condition and using the change of the acceleration (2). Further divided into a falldown state and a normal state (Normal). Since the second reversal time often falls to the ground within one second, the present invention uses the acceleration ^ (3) to judge the fall of the ward, and further accelerates the acceleration (2) sampling. Make a judgment. For example, the 6th curve (L) is the acceleration before and after the fall (2) with #又·中μ,+>,

值、二 =:::=每—二::: y' % Ms賴區間下,依實驗揭太、 門檻值"為3.55g到 ^跌倒判定 當、、跌倒判定門檻值„ g之,’係敬為-般跌倒; u火门孤值超過3 62g時 跌下、車禍撞飛、...'此疋攸如高樓 發明較佳係設定該、、跌^為嚴重的意外傷害。因此,本Value, two =:::=Every-two::: y' % Under the Ms lag interval, according to the experiment, the value of the threshold is "3.55g to ^, the fall is judged, and the fall threshold is „g,' The system is like a general fall; u, when the solitary value of the fire door exceeds 3 62g, it falls, the car crashes into the air, ... 'This is why the high-rise invention is better to set this, and drop ^ is a serious accidental injury. ,this

加速度(⑽、、物趣 仍需匈定為跌倒。 』疋门祕(3抑以上)時Acceleration ((10), the interest of interest still needs to be determined as a fall. 疋 秘 secret (3 above)

12 — 201101241 更詳言之,該跌倒條件满步驟S2 ρ 5圖所示之步驟S21至步驟s28,其中: 刀如第 「/驟切係為判斷是否為第一次感測,如判斷為 疋」’係重新回到先前之步驟S11 ;如判斷為「 一 執行步驟S22。 马否」,則 ^鄉係將先前搜集、、狀 ―態資訊“二二儲=12 — 201101241 In more detail, the fall condition is full of steps S21 to s28 shown in the step S2 ρ 5 , wherein: the knife is “/the chopping is to determine whether it is the first sensing, if it is judged as 疋"Return to the previous step S11; if it is judged as "one execution step S22. Ma no", then ^ township will collect the previous, state-state information "two two storage =

Ο it :2可隨時分析該三軸加速規11所搜集的 狀心貝訊,並執行步驟S23。 目前已進入、、計時狀態〃(詳參 」’係執行步驟S24 ;如判斷為 步驟S23係判斷是否 後續說明);如判斷為「否 「是」’則執行步驟S26。 步驟S24係藉由先前暫存之、、狀態資訊判斷是否 達到跌倒條件,當加速度(3)超過''跌侧定門權值„ 杏以亡)時,係達到跌倒條件,並執行步驟S25 ; :加速度“)未超過、、跌倒判定門檻值夕(3 %以上) 則判斷被監護者未制跌娜件,並重新_先前之 步驟S11。 步驟S25係進入、、計時狀態'以便重新執行步驟 S23 % ’可供判斷進入步驟伽’進而瞭解被監護者是否 能自行站立且繼續活動;另外,當執行步驟似後,同樣 回到先前之步驟S11。 步驟S26係贿被監護者是否能自行站立,並依據 先前暫存之、、狀«訊、行卿作#,舉例而言,該分 析皿控單70 12可利用先前暫存之人體座㈣三輛(χ —13 — 201101241 軸、y轴及z轴)與慣性座標的三抽(χ,輪 軸彼此之間的角度變化,用以觸被監護 站立;如判斷為「是」,係執行步驟S27疋否;;自订 「否」,則執行步驟S28。 如判斷為 ㈣S27料除、'計時狀態&quot;(包含累計時間 動作)’代表被監護者可自杆姑古 ^ y ' ' 報步•並重新二二=進行該緊急通 監護者之狀態。 便繼績檢測被 步驟28係判斷是否未達累計時間, 實際需求進行設定,例如匕十_ = 時間可根據醫療專業人員對於被監護人之狀ί來=累!* 此,叙设確認被監護者無法自行站立後,仍可進牛被错 :監護者無法自行站立的時間是否未達二 仃站立的時間未達預設累計時間,代表被 :: 情形仍不嚴重,故重新回到先前之步驟倒 站立的時間已達預雙翠斗 ’ ”,、法自行 驟S3。 …_ ’❹須執行急通報步 _上述步驟S21至步驟微 速;⑷是否超過、、跌倒判定Η檻值 ====:例如當被監護者曲身活動或短暫 自行站立及是否未達‘時者5 測品質。 職跌倒狀況,以有效提升跌倒债 該緊急通報步驟S3係在藉由上述步驟流程而確認被 14 201101241 i5蔓時’即時回傳被監護者跌倒資訊予監護 °可立即掌握被監護者送醫的「黃金時間 。另外,本發明廂田_ 」 件係為由被”=2加速規之齡锁測方法的終止條 蔓者或監護者依實際需求自行終止。 ;藉由如第$圖所示之流程圖,本發明之運作 拉^較佳可區分為正常狀態、異常.計時狀態及異常-解除 狀先、等一種狀態’該三種狀態皆可以每秒Μ次的速度 行相關步驟流程,其令;Ο it : 2 can analyze the heartbeat collected by the three-axis acceleration gauge 11 at any time, and execute step S23. At present, the time has elapsed, and the time is 〃 (detailed step </ RTI> step S24; if it is determined that step S23 is for subsequent explanation); if the determination is "NO", step S26 is executed. Step S24 is to determine whether the fall condition is reached by the previously stored state information, and when the acceleration (3) exceeds the 'falling side threshold weight „ apricot to die, the fall condition is reached, and step S25 is performed; : Acceleration ") does not exceed, and the fall threshold value (more than 3%) determines that the ward has not made a drop, and re-steps S11. Step S25 is to enter, and the timing state 'to re-execute step S23 % 'to judge the entry step gamma' to know whether the ward can stand and continue to move on itself; in addition, when the steps are executed, the same step back to the previous steps S11. Step S26: Whether the bribe ward can stand on its own, and according to the previously temporarily stored, the word «Xun, Xing Qing Zuo #, for example, the analysis dish control list 70 12 can use the previously temporarily stored body seat (four) three The vehicle (χ -13 - 201101241 axis, y-axis and z-axis) and the three-pitch of the inertial coordinates (χ, the angle between the axles changes to each other for the watched stand; if the judgment is "Yes", proceed to step S27疋 No; Customize “No”, proceed to step S28. If it is judged to be (4) S27, “Timekeeping Status” (including cumulative time action), “representing the ward can be self-reliant ^ y ' ' step by step • And re-two====================================================================================================来来=累!* This, the confirmation confirms that the ward cannot stand on his own, and can still enter the cow. The time when the guardian can't stand on his own is not up to the second time. The standing time is less than the preset cumulative time. The representative is: : The situation is still not serious, so it is heavy Going back to the previous steps, the time to stand up has reached the pre-double Cuidou', and the law is self-descending S3. ..._ 'Do not need to perform the emergency notification step _ the above step S21 to the step micro speed; (4) whether it exceeds, the fall judgment Η槛Value ====: For example, when the ward is in a tight-fitting activity or if it is temporarily standing on its own and if it is not up to 5, the quality is measured. The position of the job falls to effectively raise the debt. The emergency notification step S3 is performed by the above steps. And when confirmed by 14 201101241 i5 vines, 'immediately returning the ward's fall information to the guardian ° can immediately grasp the "golden time of the ward to be sent to the ward. In addition, the invention of the box _" is accelerated by "=2" The termination of the age-appropriate lock test method is terminated by the buddy or the guardian according to actual needs. By the flow chart shown in Fig. 1, the operation pull of the present invention can be better divided into a normal state, an abnormal state, and a timed state. And the abnormality - the release of the first state, and the like, the state of the three states, the speed of each of the three states can be related to the flow of the steps, the order;

被監濩者在正常狀態之步驟流程依序為步驟S11、步 驟12、步驟2卜步驟22、步驟23、步驟24、......等, 再重覆上述流程。 被[邊者在異常·計時狀態之步驟流程(在步驟24判 斷達到跌倒件,且尚無法自行站立)依序為步驟犯 '步 驟12、步驟21、步驟22、步驟23、步驟%、步驟25、 步驟11、步驟12、步驟21、步驟22、步驟23、步驟26 、步驟28、··....等’再重覆上述流程。 被監濩者在異常-解除狀態之步驟流程(在異常·計時 狀態中可自行站立)依序為步驟su、步驟12、步驟21 、步驟22、步驟23、步驟26、步驟27、…等,再重覆 上述流程。 本發明應用二轴加速規之跌倒仙方法可達到如下 所述之諸多功效,其中: 本發明應用二轴加速規之跌倒仙方法,可藉由該 狀態資訊娜步驟S1同時考量被轉者紗㈣所產生 的加速度變化及、、角度變化々,以便減少該跌倒條件 —15 — 201101241 Z㈣細,相提升跌補解確率的功 德^明應用二轴加速規之跌倒制方法,可在該跌 ==”,利用計時手段進-步確認先前: ’以達到提升偵測品質之功效。 =明應用三“速規之跌·測方法,可藉由該 監護者跌倒資訊予制,以便 丄匕 隻者,進而達到降低意外傷害比率之 功效。 士上所述纟發明應用三軸加速規之 ’確實可達到減少誤判情形,並提升偵測品質及it t 傷害比料諸多功效,以供剌照襲療科技之領i 雖然本發明已利用上述較佳實施例揭示’然其並非 =以,定本發明,任何_此技藝者在不脫離本發明之精 $和關μ,相對±述實關進行各種絲與修改仍 本發明所倾之技·,,因此本剌之 附之申請專職目所狀者鱗。 ㈤田視後 ~ 16 — 201101241 【圖式簡單說明】 第1圖:用以執行本發明應用三軸加速規之跌倒偵測 方法的系統架構示意圖。 第2圖:本發明應用三軸加速規之跌倒偵測方法所應 用之三軸加速規在跌倒前後的感測方向示意圖。 . 第3圖:本發明應用三軸加速規之跌倒偵測方法所執 • 行之步驟的方塊示意圖。 苐4圖:本發明應用三軸加速規之跌倒债測方法之人 〇 體應標相對於慣性座標之角度變化示意圖。 第5圖:本發明應用三軸加速規之跌倒/f貞測方法之步 驟流程圖。 第6圖:本發明應用三轴加速規之跌倒债測方法之跌 倒判定門檻值的曲線圖。 【主要元件符號說明】 Q 〔本發明〕 11三軸加速規 12分析監控單元 —17 —The process flow of the inspector in the normal state is sequentially step S11, step 12, step 2, step 22, step 23, step 24, ..., etc., and the above process is repeated. Step [Step 12, Step 21, Step 22, Step 23, Step %, Step 25 are performed in the sequence of steps in the abnormality and timing state (the step of determining that the fall is reached in step 24, and it is not possible to stand on its own). Step 11, Step 12, Step 21, Step 22, Step 23, Step 26, Step 28, ....., etc. 'Repeat the above process. The step of the step of the abnormality-release state (the self-standing in the abnormality/timekeeping state) is sequentially step su, step 12, step 21, step 22, step 23, step 26, step 27, ..., etc. Repeat the above process. The method for applying the two-axis accelerometer to achieve the following effects can be achieved, wherein: the method for applying the two-axis accelerometer of the present invention can be generated by the state information step S1 while considering the yarn (4). Acceleration changes, and angle changes 々, in order to reduce the fall condition - 15 - 201101241 Z (four) fine, the merits of the improvement of the rate of failure to solve the problem of the accuracy of the application of the two-axis accelerometer fall system, can be used in the fall ==", use Timing means to step forward to confirm the previous: 'To achieve the effect of improving detection quality. = Ming application 3" speed gauge drop measurement method, by the guardian fall information to the system, in order to smash the only one, and then achieve lower The effect of the accidental injury rate. The above-mentioned invention uses the three-axis accelerometer to ensure that the false positives can be reduced, and the detection quality and the damage of the ITT are improved, so that the invention can be utilized. The preferred embodiment reveals that it is not intended to be the invention, and that any skilled person can perform various techniques and modifications without departing from the essence of the present invention. Therefore, it is attached to the scale of the application for the full-time job. (5) After the field view ~ 16 — 201101241 [Simplified description of the figure] Fig. 1 is a schematic diagram of a system architecture for performing the fall detection method of the three-axis accelerometer according to the present invention. Fig. 2 is a schematic view showing the sensing direction of the three-axis accelerometer applied to the fall detection method of the three-axis accelerometer before and after the fall. Fig. 3 is a block diagram showing the steps performed by the fall detection method of the three-axis accelerometer according to the present invention.苐4Fig.: The person who applies the three-axis accelerometer's fall debt measurement method according to the present invention is a schematic diagram of the angle change of the body yoke relative to the inertial coordinate. Fig. 5 is a flow chart showing the steps of the fall/f test method of the three-axis accelerometer according to the present invention. Fig. 6 is a graph showing the threshold value of the fall determination threshold of the fall-down debt measurement method using the triaxial accelerating gauge of the present invention. [Main component symbol description] Q [Invention] 11 three-axis acceleration gauge 12 analysis monitoring unit — 17 —

Claims (1)

201101241 七、申請專利範圍: 卜-種應用三軸加速規之跌_測方法,包含: -狀態資訊擷取步驟,係藉由至少—個三軸加速規搜 #被監護者因雜㈣改變難生·速度變化及角 度變化,以獲得被監護者之狀態資訊; 一f倒條件判斷步驟’係藉由-分析監控單元判斷該 狀態資訊之加速度變化是否超過一跌倒判定門植值, 該跌倒判定門榧值為3.55g以上,且利用該狀態資訊 之角度4化判斷被監護者是^能自行站立,並在判斷 無法自行站立後進入—計時狀態,而判斷能自行站立 後則重新回到該狀態資訊擷取步驟;及 -緊急通報步驟,係域靖雜態之結果決定是否 回傳被監護者的跌倒資訊。 2、 依申料纖圍第丨項所述之應用三軸加速規之跌倒 侦測方法,其中该狀態資訊之加速度變化的轉換方式 ,係藉由接收該三軸加速規之三軸的加速度分量,以 j用乞式轉換為加速度(^,該( 龙,,)係刀別代表二軸加速規之三軸所接收的加速 度分量(g值)。 3、 依申請專利範圍第2項所述之應用三軸加速規之跌倒 偵測方法,其中該角度變化係以如下所述之公式計算 —少= α^| 其中設定慣性座標之三軸為x,軸、γ,軸及ζ,軸,該 —18 — 201101241 :二指地心引力的三軸方向’設定人體座標之 速規的三輔方向體越係指該三轴加 % Y iiik v,D °^糸為X軸與X,軸的夾角,β係 ·: ^ 轴的夾角,r係為ζ軸與ζ,轴的夾角 圍第1、2或3項所述之應用三軸加速規 Ο ο 今狀離次^法,其中該跌倒條件判斷步驟係在執行 =二,取步驟後’判斷是否為第一次感測,如 斷為;疋」’係重新_該狀態資訊擷取步驟,如判 斷為否」係儲存該狀態資訊。 5二範圍第4項所述之應用三轴加速規之跌倒 却m’其中該跌倒條件判斷步驟在儲存該狀態資 ^糸”]斷目刚是否為計時狀態,如判斷為「否」, 係錯由先前儲存之狀態資訊判斷是否達到跌倒條件, 是」,係藉由先前儲存之該狀態資訊判斷被 &amp;濩者是否能自行站立。. …依申請專鄕圍第5項所述之應用三轴加速規之跌倒 制方法’其中該跌倒條件判斷步驟在判斷被監護者 能自行站立後,係清除該計時狀態,且無須進行該緊 急通報步驟,並重新回到該狀態資訊擷取步驟。 、依申請專利第5項所述之制三轴加速規之跌倒 摘測方法,其中該跌倒條件判斷步驟在判斷被監護者 無法自行站立後,係進一步判斷是否未達一累計時間 ,如無法自行站立的時間未達該累計時間,係重新回 到該狀態資訊擷取步驟,如無法自行站立的時間已達 19 7 201101241 該累計時間,係執行 卜依陶利範圍第5項:=:步鄉。 偵測方法,其中兮膝 /用二軸加速規之跌倒 達到跌倒條件後:係進二該 狀態資訊擷取步驟。 備態,並重新回到該 =利ί? 5項所述之應用三軸加速規之跌倒 =方去,其中該跌倒條件判斷步驟在_被監護者 、到跌倒條件後’係重新回到該狀訊擷取步驟 —20 ——201101241 VII. Scope of application for patents: The application of the three-axis accelerometer to the measurement method includes: - the state information acquisition step, by at least a three-axis acceleration gauge search # Guardian due to miscellaneous (four) change is difficult · speed change and angle change to obtain status information of the ward; a f-condition judgment step 'determines whether the acceleration change of the status information exceeds a fall determination threshold value by the analysis monitoring unit, the fall determination gate The 榧 value is 3.55g or more, and the angle of the state information is used to determine that the ward is able to stand on its own, and enters the chronograph state after judging that it cannot stand on its own, and judges that it can stand on its own and then return to the state. The information acquisition step; and - the emergency notification step, the result of the domain of the miscellaneous state determines whether to return the ward's fall information. 2. The method for detecting a fall of a three-axis accelerometer according to the item mentioned in the section of the claim, wherein the change of the acceleration of the state information is obtained by receiving the acceleration component of the three axes of the three-axis accelerometer j is converted into acceleration by 乞 (^, the (龙,,) is the acceleration component (g value) received by the three axes of the two-axis acceleration gauge. 3. Application according to item 2 of the patent application scope The fall detection method of the axis accelerometer, wherein the angle change is calculated by the formula as follows: less = α^| wherein the three axes of the inertial coordinate are set to x, the axis, the γ, the axis and the ζ, the axis, the - 18 - 201101241: The three-axis direction of the two-point gravity gravitation 'the three-direction direction of the speed gauge of the human body coordinates means that the three-axis plus % Y iiik v, D ° ^ 糸 is the angle between the X-axis and the X-axis, β Department:: ^ The angle of the axis, r is the ζ axis and ζ, and the angle of the axis is applied in the first, second or third term. The three-axis acceleration rule is used. After executing step = 2, after taking the step, 'determine whether it is the first sensing, such as "断" is re- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ If the step is to store the status, the status of the condition is "time". If the determination is "No", it is determined by the previously stored status information whether the fall condition is reached, "", by the previously stored status. The information judges whether or not the person can stand on his or her own.. ... according to the application for the fall of the three-axis accelerometer method described in item 5, wherein the fall condition determination step determines that the ward can stand on his own, The timing state is cleared, and the emergency notification step is not required, and the state information acquisition step is returned to the state. The method for determining the fall condition of the three-axis acceleration gauge according to the fifth application of the patent application, wherein the fall condition is judged After judging that the ward is unable to stand on its own, it is further judged whether it has not reached a cumulative time. If the time when it is unable to stand on its own does not reach the accumulated time, it is returned. To the status information acquisition step, if the time is not able to stand on their own, the time has reached 19 7 201101241. The cumulative time is the fifth item of the Buy Taoli range: =: step township. Detection method, in which the knees/two axes are used. After the fall of the acceleration gauge reaches the fall condition: the second step of the state information acquisition step is taken. The state is returned, and the fall of the application of the three-axis acceleration gauge described in item 5 is determined, and the fall condition is judged. The steps are in the _ ward, after the fall condition, 'return to the message acquisition step - 20 ——
TW98121994A 2009-06-30 2009-06-30 A falling down detection method implemented by tri-axis accelerometer TWI374408B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW98121994A TWI374408B (en) 2009-06-30 2009-06-30 A falling down detection method implemented by tri-axis accelerometer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW98121994A TWI374408B (en) 2009-06-30 2009-06-30 A falling down detection method implemented by tri-axis accelerometer

Publications (2)

Publication Number Publication Date
TW201101241A true TW201101241A (en) 2011-01-01
TWI374408B TWI374408B (en) 2012-10-11

Family

ID=44836960

Family Applications (1)

Application Number Title Priority Date Filing Date
TW98121994A TWI374408B (en) 2009-06-30 2009-06-30 A falling down detection method implemented by tri-axis accelerometer

Country Status (1)

Country Link
TW (1) TWI374408B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103021128A (en) * 2011-09-20 2013-04-03 英业达股份有限公司 Old people tumbling warning system and method thereof
CN113485316A (en) * 2020-03-16 2021-10-08 东莞富强电子有限公司 Movement tracking device and method thereof

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI494568B (en) * 2014-06-16 2015-08-01 Nat Inst Chung Shan Science & Technology Motion detection method and device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103021128A (en) * 2011-09-20 2013-04-03 英业达股份有限公司 Old people tumbling warning system and method thereof
CN113485316A (en) * 2020-03-16 2021-10-08 东莞富强电子有限公司 Movement tracking device and method thereof

Also Published As

Publication number Publication date
TWI374408B (en) 2012-10-11

Similar Documents

Publication Publication Date Title
WO2017049957A1 (en) Intelligent falling detection and alarming apparatus and processing method thereof
CN205041401U (en) Equipment is worn detection device and is had and wears guardianship device that detects function
CN104504854B (en) Alarming method for detecting fall over of human body
Chen et al. Wearable sensors for reliable fall detection
Kangas et al. Sensitivity and false alarm rate of a fall sensor in long-term fall detection in the elderly
WO2017049958A1 (en) Paralysis detection and alarming apparatus and processing method thereof
AU2009305075B2 (en) A fall detection system and a method of operating a fall detection system
CA2859142C (en) Method and computer program for monitoring use of an absorbent product
CN201691930U (en) Human movement state monitoring device
Yuan et al. Power-efficient interrupt-driven algorithms for fall detection and classification of activities of daily living
US20200380841A1 (en) Impact Force Estimation and Event Localization
US10930134B2 (en) Hand hygiene monitoring system and its application method
WO2016095871A1 (en) Wearable device and method therefor
TW200912814A (en) Wearable activity sensing device and method thereof
CN104408877A (en) Alarming system for detecting fall over of human body
CN105528859A (en) Nursing intelligent alarm system based on human body falling down detection technology
Chen et al. A wireless real-time fall detecting system based on barometer and accelerometer
EP1974662A1 (en) Fall detector
TW201101241A (en) A falling down detection method implemented by tri-axis accelerometer
Tong et al. A research on automatic human fall detection method based on wearable inertial force information acquisition system
Tan et al. Reliable fall detection system using an 3-DOF accelerometer and cascade posture recognitions
CN205697787U (en) Intelligence is fallen warning neck ring
Babu et al. Wearable device based fall prediction and alert mechanism for aged people using IoT technology
CN109171751A (en) A kind of human posture recognition method and device
GB2525903A (en) A device for determining unconciousness of a subject and a device for detecting a fall by a subject

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees