`US 6,614,348 B2
`(10) Patent No.:
`(12)
`Ciccolo et al.
`(45) Date of Patent:
`Sep. 2, 2003
`
`
`US006614348B2
`
`(54) SYSTEM AND METHOD FOR MONITORING
`BEHAVIOR PATTERNS
`
`(75)
`
`Inventors: Arthur C. Ciccolo, Ridgefield, CT
`(US); Phillip Hobbs, Briarcliff Manor,
`NY (US); John D. Mackay, Sleepy
`Hollow, NY (US), Howard E. Sachar,
`MountKisco, NY (US)
`International Business Machines
`Corporation, Armonk, NY (US)
`
`(73) Assignee:
`
`6/1989) Tsuji occ 340/541
`4,839,631 A *
`5,841,137 A * 11/1998 Whitney oe 250/338.5
`5,942,976 A
`8/1999 Wieseret al. 0... 340/565
`5,998,780 A
`12/1999 Kramer vvesseseeseeeees 340/555
`6,028,514 A *
`2/2000 Lemelsonet al.
`....... 340/573.1
`6,043,493 A
`3/2000 Kimetal. cssseecsssssseeee 250/349
`6,054,928 A *
`4/2000 Lemelsonetal.
`....... 340/573.4
`6,384,414 B1 *
`5/2002 Fisheret al. wee 340/567
`6,437,696 B1 *
`8/2002 Lemelson et al.
`....... 340/573.4
`
`* cited by examiner
`
`(*) Notice:
`
`Subject to any disclaimer, the term ofthis
`patent is extended or adjusted under 35
`US.C. 154(b) by 181 days.
`
`(21) Appl. No.: 09/814,785
`
`Primary Examiner—Van Trieu
`(74) Attorney, Agent, or Firm—Stephen C. Kaufman, Esq.;
`McGinn & Gibb, PLLC
`
`Filed:
`
`(22)
`65
`(65)
`
`ABSTRACT
`(57)
`Mar. 23, 2001
`Prior
`Publication Dat
`A system and method for monitoring behavior patterns
`ONION
`which effectively distinguishes between alarming and non-
`US 2002/0135484 Al Sep. 26, 2002
`(51) Unt, CL.cceccccccceccccccssseesssteeseeeseesen GO8B 13/00—alarming behavior patterns, includes at least one sensor for
`(52) US. Ch. cece 340/541; 340/565; 340/573.4__—_detecting behavior patterns, a memory device coupled to the
`(58) Field of Search 0.00.00. 340/555, 565,
`sensor,
`for storing standard behavior patterns, and a
`340/545.3, 556, 557, 573.3, 573.4, 541,
`processor, coupled to the memory device, for comparing
`576, 575; 128/903; 600/529, 543, 558,
`standard behavior patterns with detected behavior patterns,
`595; 250/338.5, 339.02, 339.05, 339.14
`and causing a response to be activated when standard
`References Cited
`behavior patterns and detected behavior patterns have a
`predetermined relationship.
`U.S. PATENT DOCUMENTS
`
`(56)
`
`ATITATT A
`
`10/1988 Watson voces. 340/573.4
`
`29 Claims, 6 Drawing Sheets
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 1
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 1
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 1 of 6
`
`US 6,614,348 B2
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 2
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 2
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 2 of 6
`
`US 6,614,348 B2
`
`q2ls
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 3
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 3
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 3 of 6
`
`US 6,614,348 B2
`
`DETECT BEHAVIOR PATTERNS
`
`COMPARE DETECTED BEHAVIOR PATTERNS
`TO STANDARD BEHAVIOR PATTERNS
`
`DOES PREDETERMINED RELATIONSHIP EXIST?
`
`NO
`
`YES
`
`HAS PREDETERMINED CONDITION
` HAS A PREDETERMINED CONDITION
`BEEN MET FOR STORING DETECTED
`BEEN MET FOR ERASING THE
`
`BEHAVIOR PATTERN AS A
`STANDARD BEHAVIOR PATTERN?
`
`
`STANDARD BEHAVIOR PATTERN?
`
`
`
`
`
`
`
`
`
`
`- STORE DETECTED BEHAVIOR PATTERN AS
`A STANDARD BEHAVIOR PATTERN
`~ STORE PREDETERMINED RELATIONSHIP
`~ STORE RESPONSE TO BE ACTIVATED
`
`
`
`
`
`ERASE STANDARD
`
`BEHAVIOR PATTERN
`
`
`FIG.2C
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 4
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 4
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 4 of 6
`
`US 6,614,348 B2
`
`FIG.3
`
`
`SENSOR 3
`
`
`SENSOR 1
`TYPE A
`
`
`
`
`
`201
` 201
`
`STORE STANDARD BEHAVIOR PATTERNS
`
`
`
`DETECT BEHAMOR PATTERNS
`
`
`
`
`
`
`
`
`
`
`
`COMPARE DETECTED BEHAVIOR PATTERNS
`TO STANDARD BEHAVIOR PATTERNS
`
`DOES A PREDETERMINED RELATIONSHIP EXIST
`
`YES
`
`ACTIVATE A RESPONSE
`
`FIG.4A
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 5
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 5
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 5 of6
`
`US 6,614,348 B2
`
`STORE STANDARD BEHAVIOR PATTERNS
`
`DETECT BEHAVIOR PATTERNS
`
`COMPARE DETECTED BEHAVIOR PATTERNS
`TO STANDARD BEHAVIOR PATTERNS
`
`DOES A PREDETERMINED RELATIONSHIP EXIST?
`
`YES
`
`
`
`
`
`
`
`
`
`HAS PREDETERMINED CONDITION
`
`BEEN MET FOR STORING DETECTED
`
`BEHAVIOR PATTERN AS A
`
`STANDARD BEHAVIOR PATTERN?
`
`
`
`HAS A PREDETERMINED CONDITION
`BEEN MET FOR ERASING THE
`
`STANDARD BEHAVIOR PATTERN?
`
`
`
`— STORE DETECTED BEHAVIOR PATTERN AS
`A STANDARD BEHAVIOR PATTERN
`
`- STORE PREDETERMINED RELATIONSHIP
`— STORE RESPONSE TO BE ACTIVATED
`
`
`
`FIG.4B
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 6
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 6
`
`
`
`U.S. Patent
`
`Sep. 2, 2003
`
`Sheet 6 of 6
`
`US 6,614,348 B2
`
`534
`
`NETWORK
`
`918
`
`(/0COMMUNICATIONS
`
`540
`
`521
`
`16
`
`14
`
`i1
`
`
`
`536
`
`522
`
`DISPLAY
`
`ADAPTER
`
`USERINTERFACE
`
`ADAPTER
`
`912
`
`524
`
`532
`
`928
`
`|
`
`926
`
`A
`
`FIG.5
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 7
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 7
`
`
`
`US 6,614,348 B2
`
`1
`SYSTEM AND METHOD FOR MONITORING
`BEHAVIOR PATTERNS
`
`BACKGROUND OF THE INVENTION
`
`1. Field of the Invention
`
`The present invention generally relates to a monitoring
`system and, more particularly, a system for monitoring
`behavior patterns of individuals and other animated objects
`which effectively distinguishes between alarming and non-
`alarming patterns.
`2. Description of the Related Art
`As shownin FIG. 1, conventional monitoring systemsare
`often audiovisual systems 100 that employ audio and video
`equipment to monitor individuals. In such systems, micro-
`phones 101 and cameras 102 are used to detect behavior
`patterns (e.g., a monitored person lying in a bed between
`10:00 p.m. and 7:00 a. m., showering between 7:30 a.m. and
`8:00 a.m., etc) as they occur. Signals representing such
`patterns are transmitted from the microphones and cameras
`to a video display 103 and speakers 104 which are monitored
`continuously by a human monitor 105. When the human
`monitor 105 observes an alarming behavior pattern (e.g., the
`monitored person lying on the floor, the monitored person
`not in bed at midnight, etc.), the human monitor may take
`corrective action (e.g.,
`tend to an elderly person whose
`behavior patterns are being monitored)or report such alarm-
`ing pattern to the appropriate person or agency (e.g. a nurse
`whocan tend to such elderly person).
`However, such audiovisual systems are intrusive, ineffi-
`cient and costly. For example, a video signal requires
`substantial bandwidth making it burdensomeand costly to
`transmit. Further, the use of a human monitor is costly and
`subject to human error. Human monitors mustbe trained and
`must remain in close proximity to the monitored individual.
`In the case of video observation, the low rate of alarming
`events (visual changes in the scene) often lead to poor
`attention by support people and the possibility of ignoring an
`alarming event. Therefore, if the human monitoris not active
`and vigilant, an alarming pattern can be easily overlooked.
`Furthermore, such systems are unnecessarily intrusive into
`the lives of the persons being monitored because the human
`monitor’s observations are not
`limited to just alarming
`behaviorpatterns, but must include each and every action of
`the person being monitored.
`Other conventional monitoring systems include motion
`sensing systems which use motion sensors to detect move-
`ment in a space being monitored. Motion sensors are typi-
`cally photosensors that detect moving objects based on
`discrete approximations of space or time. In such systems,
`the sensors are connected to an alarm circuit which typically
`has an audible alarm. However, such motion sensing moni-
`toring systems monitor only a predefined space, not monitor
`behavior patterns of individuals. This severely limits the
`utility of such systems. Further, Generally, such systems do
`not distinguish between motion caused by a person and
`motion from any other entity of comparable size or with a
`comparable extent of motion. In addition, such systems do
`not distinguish between a monitored individual and a non-
`monitored individual. In either case, regardless of whether
`the individual detected is monitored or non-monitored, if
`such a system is active and functioning properly, then it will
`alarm upon the individual entering the space.
`Therefore, motion sensing systems do notdetect alarming
`behavior patterns. For example, such systems cannot moni-
`tor elderly individuals with alzheimer’s disease to detect
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`when such individuals are not in bed at a certain hour or are
`lying on the floor of their room, etc. Similarly, such systems
`cannot monitor infants to detect when such infants are not in
`
`their cribs, or are near dangerous objects such as windows
`or appliances, etc. Nor can such systems monitor ware-
`houses or retail shopping areas to detect behavior patterns
`that would indicate theft.
`
`Another conventional system is an infrared monitoring
`system which uses infrared sensors to monitor spaces such
`as museums and banks. Infrared sensors operate based on
`Stefan-Boltzmann’s law that every body radiates an energy
`proportional to a fourth powerof an absolute temperature of
`the body. Such sensors typically detect radiant energy emit-
`ted from bodies, human or otherwise, within a wavelength
`range from approximately 6 to 15 micrometers.
`A monitoring system utilizing infrared sensors is
`disclosed, for example,
`in Weiser et al.
`(U.S. Pat. No.
`5,942,976). Infrared sensors used in such systems include a
`housing with an entrance window whichistransparentto the
`infrared radiation, focusing optics, one or more infrared
`sensors, and an electrical signal evaluation circuit. Such
`systems further include an alarm circuit which typically has
`an audible alarm.
`
`if an intruder enters the space
`With such a system,
`monitored by the infrared detector, his infrared body radia-
`tion enters through the entrance window into the detector
`and is focused by the focusing optics onto the infrared
`sensors. The infrared sensors output a signal to the circuit
`which amplifies the signal and compares it to a predeter-
`mined threshold. If the threshold is exceeded,
`then an
`intrusion alarm signal is generated.
`However, infrared systems also have their shortcomings.
`For example, like other monitoring systems, existing infra-
`red systems either detect an alarming event or they don’t.
`These systems typically provide no additional information
`(e.g., duration, specific location, frequency of occurrence,
`etc.) about the event. In other words, these systems have no
`ability to interpret a pattern of behavior and select from a
`variety of potential responses.
`infrared systems, similar to
`In addition, conventional
`motion sensing systems, monitor only a predefined space,
`not behavior patterns of individuals. Thus,
`like motion
`sensing systems, such infrared systemsare limited in utility.
`For example, such systems cannot distinguish between a
`monitored individual and a non-monitored individual. In
`
`is
`regardless of whether the individual
`either case,
`monitored,
`if such a system is active and functioning
`properly, it will alarm uponthe individual entering the space.
`Furthermore, because conventional infrared systems have
`simple and nondiscriminating detectors, the systems often
`detect events that aren’t actually alarming andare, therefore,
`result in a high false/positive responserate.
`Therefore, as with other systems, conventional infrared
`systems cannot effectively monitor behavior patterns and
`detect alarming behavior patterns that would indicate, for
`example, theft in a warehouse orretail store, or a potential
`harm to an infant or an alzheimer’s patient.
`SUMMARYOF THE INVENTION
`
`is,
`it
`In view of the foregoing and other problems,
`therefore, an object of the present invention to provide a
`system and method for monitoring behavior patterns of
`individuals which effectively distinguishes between alarm-
`ing and non-alarming behavior patterns.
`In a first aspect, a system for monitoring behaviorpatterns
`includes sensors for detecting behavior patterns, a memory
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 8
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 8
`
`
`
`US 6,614,348 B2
`
`3
`device for storing behavior patterns, a processor for com-
`paring detected behavior patterns with standard behavior
`patterns and activating a response when the detected behav-
`ior pattern and at least one standard behavior pattern have a
`predetermined relationship, such as when the detected data
`matches the stored data, or which the detected data differs
`from stored data.
`
`The system may include a plurality of sensors which are
`interconnected and the sensors may be infrared sensors for
`detecting infrared radiation. For example, the infrared sensor
`may detect a variation in radiant energy of less than one
`Kelvin. The system may use the multiple sensors to detect
`behavior patterns comprised of sequences of patterns, com-
`pare the detected behavior patterns to the standard behavior
`patterns, generate a signal to activate additional sensors to
`detect supplementary data, and transmit this supplementary
`data to a remote location.
`
`The memory device may be a conventional semiconduc-
`tor memory device. Further, the processor may be an adap-
`tive processor programmed with a learning algorithm so that
`the system “learns” new standard behavior patterns while it
`operates and “forgets” old standard behavior patterns that
`may no longer be considered alarming.
`The response activated may also include additional sen-
`sors for collecting additional information. The response may
`also include a human response, an audiovisual or photo-
`graphic device or an auto-dialer which makes a call to the
`police, ambulance,etc.
`In a second embodiment, a method of monitoring behav-
`ior patterns includes storing standard behavior patterns,
`detecting behavior patterns, comparing detected behavior
`patterns to standard behavior patterns, and activating a
`response when a detected behavior pattern and at least one
`standard behaviorpattern have a predeterminedrelationship,
`such as when the detected data matches the standard data, or
`which the detected data differs from standard data.
`
`The inventive method may also employ multiple sensors
`to detect behavior patterns comprised of sequences of
`patterns, compare the detected behavior pattern data to the
`standard behaviorpatterns, and activate a response that may
`include additional sensors to detect supplementary data, and
`transmit this supplementary data a remote location.
`the
`With the novel features of the claimed invention,
`behavior patterns of individuals can be monitored with an
`improved ability to distinguish between alarming and non-
`alarming behaviorpatterns. In addition, based on the behav-
`ior pattern detected, the claimed invention may initiate a
`variety of responses, such as the collection of additional data
`from a plurality of heterogeneous sensors. Moreover, the
`system at the device level and at the aggregation of devices
`level, can discover patterns that should be categorized as
`alarming (or normal) and incrementally alter what condi-
`tions precipitate the transmission of an alarm or other
`information.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`The foregoing and other objects, aspects and advantages
`will be better understood from the following detailed
`description of a preferred embodimentof the invention with
`reference to the drawings, in which:
`FIG. 1 illustrates a conventional monitoring system 100
`using audio/video equipment;
`FIG. 2Aillustrates a monitoring system 200 according to
`a preferred embodimentof the present invention;
`FIG. 2B illustrates exemplary behavior patterns that may
`be detected by the system 200;
`
`10
`
`15
`
`20
`
`25
`
`30
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`4
`FIG. 2C illustrates a learning algorithm that may be used
`by the processor 204 to automatically store new standard
`behavior patterns and erase existing standard behaviorpat-
`terns;
`FIG. 3 illustrates a monitoring system 200 having mul-
`tiple sensors, according to an aspect of a preferred embodi-
`ment of the present invention;
`FIG. 4A is a flow chart illustrating a method for moni-
`toring behavior patterns according to a second embodiment
`of the present invention;
`FIG. 4B is a flow chart illustrating a method for moni-
`toring behavior patterns which includes a learning
`algorithm, according to a second embodimentof the present
`invention;
`FIG. 5 illustrates an hardware/information handling sys-
`tem 500 for incorporating the present invention; and
`FIG. 6 illustrates a signal bearing medium 600 (ic.,
`storage medium) for storing steps of a program of a method
`according to the present invention.
`
`DETAILED DESCRIPTION OF THE
`PREFERRED EMBODIMENT OF THE
`INVENTION
`
`Referring now to the drawings, FIG. 2A illustrates a
`monitoring system according to a preferred embodiment of
`present invention.
`In a preferred embodiment, a monitoring system 200
`includes one or more sensors 201 for detecting radiant
`energy levels which define behavior patterns of monitored
`objects. The sensors 201 transmit data when there is a
`changein the field of view of the sensors 201. Such sensors
`may be, for example, infrared sensors. Further, the sensors
`201 may be inexpensive and highly sensitive, and may be
`located at any location that is not prominent and allow the
`sensors 201 to operate without interference, such as the
`ceiling or wall of a room. In addition, since only alarms
`and/or small amounts of imagedataaresent, little bandwidth
`and therefore, little wiring is needed.
`In addition, the sensors 201 are easy to install and are
`programmedso that they understand the geography of the
`space within which they operate. The highly sensitive sen-
`sors 201 also allow the inventive system 200 to efficiently
`monitor a large space. For example, if each pixel in a sensor
`201 provides one byte of data and images one square foot of
`area in the monitored field, and the embodied array is 8
`feetx12 feet, a complete representation of the monitored
`field is under 100 bytes, uncompressed. Moreover, depend-
`ing on the granularity of the data required, the area covered
`by one pixel can be adjusted by remotely adjusting the lens.
`As further shown in FIG. 2A, the inventive system 200
`includes a memory device 203 connected to the sensors 201.
`The memory device 203 may be a conventional memory
`device (e.g., random access memory (RAM)) andis used to
`store information,
`including standard behavior patterns
`whichare defined by radiant energy level data. For example,
`if the system 200 is being used to monitor an alzheimer’s
`patient, the standard behavior pattern stored in the memory
`device 203 may be defined by radiant energy level data
`corresponding to the patient’s energy level existing at a
`certain location at a certain hour.
`
`FIG. 2B illustrates two exemplary behavior patterns 250,
`260 which are defined by radiant energy levels. As shown in
`FIG. 2B, a first behavior pattern 250 is defined by radiant
`energy level data corresponding to an individual entering a
`door 251, dwelling for a certain time period at points A, B,
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 9
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 9
`
`
`
`US 6,614,348 B2
`
`5
`C and D, then exiting door 251. Similarly, a second behavior
`pattern 260 is defined by radiant energy level data corre-
`sponding to an individual entering a door 262, dwelling for
`a certain time period at points E, F, G and H,then exiting
`door 263.
`
`Any behavior pattern which can be defined by radiant
`energy level data can be stored in the memory device. For
`example, an employee moving repeatedly between a locker
`room and a work station may be a behavior pattern indicat-
`ing a theft by the employee. Radiant energy level data
`defining this behavior pattern may be stored in the memory
`device 203 as a standard behavior pattern. Similarly, a retail
`store customer making repeated movements into and out of
`a handbag or an individual moving repeatedly into and out
`of the store without making a purchase, may be a behavior
`pattern indicating shoplifting. In that case, radiant energy
`level data defining such a suspicious behavior pattern may
`be stored in the memory device 203 as a standard behavior
`pattern.
`The system further includes a processor 204 which is
`coupled to the sensors 201 and to the memory device 203
`and which compares detected behavior patterns with the
`standard behavior patterns which are stored in the memory
`device 203. The processor 204 may be a conventional
`microprocessor. Whenit is determined by the processor 204
`that the detected behavior patterns and the standard behavior
`patterns have a predeterminedrelationship (e.g., the detected
`data matches the standard data or the detected data differs
`from the standard data), the processor 204 may cause an
`alarm signal to be generated.
`Further, when he processor 204 finds a predetermined
`relationship exists between a detected behavior pattern and
`a standard behaviorpattern, the processor causes a response
`206 to be activated. The response 206 shown in FIG. 2A is
`an audiovisual system, however, the inventive system may
`include any variety of responses 206. For example,
`the
`response 206 may include a humanresponseor an audible
`alarm suchas a siren or a visual alarm such asflashing lights.
`In addition, the response 206 may include other data capture
`devices such as a camerafor taking still photographs or an
`audiovisual display that can be observed by a human moni-
`tor. The response 206 may also include an auto-dialer for
`automatically dialing the police, ambulance orfire depart-
`ment.
`
`the inventive system 200 may include a
`In addition,
`variety of potential responses 206, and may activate a
`particular response 206 based on the standard behavior
`pattern for which a predetermined relationshipis identified.
`For example, if the system is being used in a retail store, and
`the standard behavior pattern includes an individual entering
`and leaving the store three times in the same day without
`making a purchase which would indicate shoplifting, the
`response 206 activated may include an audiovisual system
`which follows the individual throughout the store the next
`time he enters. On the other hand, if the standard behavior
`pattern includes an intense heat in the store indicatinga fire,
`the response 206 activated may include an automatic phone
`call to the fire department. The particular response 206 to be
`activated may be stored in the memory device 203.
`The inventive system 200 mayalso store in the memory
`device 203, information regarding the a history of detected
`behavior patterns, predetermined relationships identified,
`responsesactivated and other important information such as
`the date, time and comments from a human monitor regard-
`ing a response 206 activated.
`The inventive system 200 may include a controller 207
`for controlling the system 200. The controller 207 may be
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`6
`hardwired to the system or may be wirelessly connected to
`the system 200. For example, the controller 207 may be a
`personal computer and include a keyboard for inputting and
`modifying data (e.g., standard behavior patterns, predeter-
`mined relationship which causes an alarm signal
`to be
`generated, and the type of response activated) to the memory
`device 203. The controller 207 may include a video display
`unit for visually displaying data stored on the memory
`device 203. The controller 207 may also be used for such
`functions as activating and deactivating the system 200 and
`controlling the functions of the processor 204 such as
`overriding the responseactivated (e.g., turning off an audible
`alarm or audiovisual system).
`Further, the processor 204 of the inventive monitoring
`system mayinclude a learning function so thatit can “learn”
`new standard behavior patterns and erase (ie., “forget) old
`behavior patterns from the list of standard behavior patterns
`in the memory device. FIG. 2C showsa learning algorithm
`270 that may be used by the processor 204. As shownin FIG.
`2C, after the system determines that no predetermined
`relationship exists,
`the processor 204 may determine if
`certain predetermined conditions have been metfor storing
`the detected behavior pattern as a new standard behavior
`pattern. For instance, the predetermined condition may be
`that the detected behavior pattern has been detected n times
`over a time period t. If the processor 204 determinesthat this
`condition has been not been met, the algorithm returns the
`system 200 back to detecting behavior patterns. However, if
`this predetermined condition in the learning algorithm 270 is
`met, the processor 204 causes the detected behavior pattern
`to be automatically stored in the memory device as a new
`standard behavior pattern, and automatically causes to be
`stored a pretermined relationship corresponding to the new
`standard behavior pattern and a response to be activated
`when the predetermined relationship exists.
`On the other hand,if the processor 204 determinesthat a
`predetermined relationship does exist, the processor may
`then determine if a predetermined condition has been met
`for erasing the standard behavior pattern having a predeter-
`minedrelationship with the detected behavior pattern. For
`instance, such a predetermined condition may be that the
`system 200 has detected this particular predetermined rela-
`tionship x times over a time periodt. If the processor 204
`determines that the predetermined condition has not been
`met, the processor 204 proceeds to cause an alarm signal to
`be generated. On the other hand,
`if the predetermined
`condition has been met, the processor 204 causes the cor-
`responding standard behavior pattern to be erased from the
`memory device 203.
`Furthermore, as shown in FIG. 3, the inventive system
`200 may include a plurality of sensors 201 coupled to the
`processor 204. The sensors 201 may have varying specifi-
`cations as shown in FIG. 3. The sensors 201 may work
`together to detect behavior patterns comprised of sequences
`of patterns and the processor 204 compares the behavior
`patterns detected by the plurality of sensors to the standard
`behavior patterns stored in the memory device 203. If a
`predetermined relationship is found, the response 206 acti-
`vated may include additional sensors 201 to detect supple-
`mentary data which may be wirelessly transmitted to a
`remote location 390 (e.g. the controller 207) to be analyzed.
`The inventive monitoring system 200 also provides an
`effective and efficient system of monitoring the behavior
`patterns 202 of individuals such as hospitalpatients, infants,
`retail customers, employees and prisoners. The sensors 201
`can detect whether an individual whose behavior patterns
`are being monitored has an elevated or depressed body
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 10
`
`IPR2018-01093
`Apple Inc. Ex. 1010 Page 10
`
`
`
`US 6,614,348 B2
`
`10
`
`15
`
`20
`
`25
`
`30
`
`7
`temperature. The system may also detect alarming events
`surrounding such a monitored individual such asfire, low/
`high temperature, smoke, etc. and whether windows or
`doors are open or closed. Thus, there may be no need for
`attendants or continuous live video or audio. Therefore, the
`system saves moneyandis not subject to humanerrorlike
`the audiovisual monitoring system. In addition, this system
`is not
`intrusive like conventional audiovisual systems
`because the behavior patterns of the individuals being moni-
`tored are not continuously monitored by another person.
`Further, unlike motion sensor and conventional
`infrared
`sensor monitoring systems, the inventive system is able to
`distinguish between a monitored individual and a non-
`monitored individual by comparing each individual’s behav-
`ior patterns to predetermined patterns. Moreover, the system
`requires no morefor installation than a smoke detector and
`has a short calibration procedure.
`FIG. 4 provides a flow chartillustrating a method 400 for
`monitoring behavior patterns according to a second embodi-
`ment of the present invention.
`According to the claimed method 400 of monitoring
`behavior patterns 202 (see, also, FIGS. 2C and 3), standard
`behavior patterns are stored (410). This may be performed
`by a memory device such as a semiconductor memory
`device or, more specifically, a conventional semiconductor
`RAM. As explained above, the standard behavior patterns
`may be defined by radiant energy level data.
`The inventive method 400 further includes detecting
`(420) behavior patterns by detecting radiant energies. As
`explained above, this may be performed by more than one
`sensor 201 such as infrared sensors.
`
`8
`new standard behaviorpattern and a responseto be activated
`when the predetermined relationship exists, are also auto-
`matically stored (450).
`Onthe other hand,if it is determined that a predetermined
`relationship does exist, it is then determined if a predeter-
`mined condition has been met for erasing the standard
`behavior pattern having a predetermined relationship with
`the detected behavior pattern. If it is determined that the
`predetermined condition has not been met, a responseis then
`activated (440). On the other hand, if the predetermined
`condition has been met, the corresponding standard behavior
`pattern is erased from the memory device (460).
`FIG. 5 illustrates a typical hardware configuration of an
`information handling/computer system in accordance with
`the invention and which preferably hasat least one processor
`or central processing unit (CPU) 511.
`The CPUs511 are interconnected via a system bus 512 to
`a random access memory (RAM) 514, read-only memory
`(ROM) 516, input/output (I/O) adapter 518 (for connecting
`peripheral devices such as disk units 521 and tape drives 540
`to the bus 512),user interface adapter 522 (for connecting a
`keyboard 524, mouse 526, speaker 528, microphone 532,
`and/or other user interface device to the bus 512), a com-
`munication adapter 534 for connecting an information han-
`dling system to a data processing network, the Internet, an
`Intranet, a personal area network (PAN),etc., and a display
`adapter 536 for connecting the bus 512 to a display device
`538 and/or printer 539 (e.g., a digital printer or the like).
`In addition to the hardware/software environment
`described above,a different aspect of the invention includes
`a computer-implemented method for performing the above
`method. As an example, this method may be implemented in
`The inventive method 400 further includes comparing
`the particular environment discussed above.
`(430) detected behavior patterns with standard behavior
`Such a method may be implemented, for example, by
`patterns which are stored. This may be performed by a
`operating a computer, as embodied by a digital data pro-
`processor as explained above. If no predeterminedrelation-
`cessing apparatus,
`to execute a sequence of machine-
`ship between the detected behavior pattern and the standard
`readable instructions. These instructions may reside in vari-
`behavior patterns is found, no responseis activated. If such
`ous types of signal-bearing media.
`a predetermined relationship is found, a response is acti-
`vated (440). As explained above, the response may include
`Thus, this aspect of the present invention is directed to a
`programmedproduct, comprising signal-bearing media tan-
`a human response, an audiovisual or still photographic
`gibly embodying a program of machine-readable instruc-
`capture device or to an auto-dialer which causes a telephone
`
`call to be initiated to the proper authorities. tions executable byadigital data processor incorporating the
`CPU 511 and hardware above, to perform the method of the
`Furthermore,
`the inventive method 400 may use the
`invention.
`multiple sensors 201 coupled to a processor 204. The
`This signal-bearing media may include, for example, a
`sensors 201 may have varying specifications as shown in
`RAM contained within the CPU 511, as represented by the
`FIG. 3. The sensors 201 may work together to detect
`fast-access storage for example. Alternatively, the instruc-
`behavior patterns comprised of sequencesof patterns and the
`tions may be contained in another signal-bearing media,
`processor 204 comparesthe detected behavior patternsto the
`such as a magnetic data storage diskette 600 (FIG. 6),
`standard behavior patterns. The processor 204 may then
`directly or indirectly accessible by the CPU 511.
`generate a signal to activate additional sensors 201 to detect
`supplementary data, and transmit this supplementary data a
`Whethercontained in the diskette 600, the computer/CPU
`remote location 390.
`511, or elsewhere, the instructions may be stored onavariety
`of machine-readable data storage media, such as DASD
`In addition,
`the claimed method 400 may include a
`storage (e.g., a conventional “hard drive” or a RAID array),
`learning algorithm (e.g., see FIG. 2C) so as to include
`magnetic tape, electronic read-only memory (e.g., ROM,
`“learning” new standard behavior patterns and erasing (i.e.,
`EPROM, or EEPROM), an optical storage device (e.g.
`forgetting) old behavior patterns from the list of standard
`CD-ROM, WORM,DVD, digital optical tape, etc.), paper
`behavior p