JP2010131280A - Method and apparatus for assisting mental state determination - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and an apparatus for recognizing and determining the mental state, psychology or psychological state of a writer by quantifying temporal and spacial information on hand-written letters and pictures and extracting the quantity of characteristics. <P>SOLUTION: The apparatus includes a digital pen connecting device 101 for acquiring stroke data from a digital pen 111; a recognition part 103 for recognizing the written letters or pictures from the stroke data; and an arithmetic operating part 102 for comparing and evaluating the recognized letters or pictures. The state of the writer is determined by a step 121 for measuring information on the hand-written letters; a step 122 for recording the measured information on the written letters; a step 123 for clipping and identifying the letters or pictures based on the information on the written letters; a step for computing the quantity of characteristics from the stroke information by the clipped and identified letters or pictures as units; and a step 124 for displaying the quantity of characteristics. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a method and apparatus for grasping a writer's mental / psychological / physiological state by quantifying spatiotemporal information of handwritten characters and pictures and extracting feature quantities.

"Writing / drawing" is intended to convey the written content to others. Therefore, conventionally, only the written characters and the contents of the drawn pictures have attracted attention. However, “writing / drawing” is executed under the control of the brain. It is the brain that intends and decides what is written and drawn, that is, what and how to draw. The brain also controls the movement of tools such as pencils and ballpoint pens that “write and draw” with their hands. For this reason, written characters and drawn pictures reflect brain function, and the state of the writer's brain, that is, the mental, psychological, and physiological states are reflected.
For example, it has been reported that when depression occurs, the writing speed decreases (Non-Patent Document 1, etc.). In Parkinson's disease patients, a symptom (called dwarfism) in which the size of characters gradually decreases may appear (Non-Patent Document 2, etc.). Furthermore, the HTP test (House, tree, person test), which is one of the psychological tests, draws a house, a tree, and a person together on a single sheet, and draws the balance and size of the three sizes. A subject's mental state can be evaluated by a position etc. (nonpatent literature 3 etc.). Furthermore, children with learning disabilities can be screened at the frequency of rewriting that occurs during the drawing process by letting people draw (Figure 4).

  On the other hand, a digital pen is one of devices that digitize and acquire handwritten writing / drawing information (for example, Patent Document 1). This is a ballpoint pen that actually uses ink to incorporate a device for reading positional information on paper. The dot pattern printed on the paper is recognized by the optical system (consisting of an infrared laser, a CCD camera, etc.) built in the digital pen, and the pen tip position is calculated and identified, and recorded as time-series data. This is realized by doing (Patent Document 1, etc.).

  If the writing / drawing process can be acquired as time-series data, the input device is not limited to a digital pen, and may be a touch panel used in a tablet PC, a mouse used in PC drawing software, or the like. However, considering the position resolution and sampling rate (time resolution), for example, one stroke (corresponding to one stroke in Kanji) under the character size and speed of “natural writing / drawing in everyday situations” In order to obtain the writing speed with a resolution that can be analyzed, at least the position resolution must be 1 mm or less, and the time resolution must be 33 ms (video rate) or less. The digital pen is an effective means for realizing this at a low cost. Being realistic ensures the convenience and mobility.

JP 2006-154986 JP Kinematical analysis of handwriting movements in depressed patients.Mergl R et.al, Acta Psychiatrica Scandinavia, 109 109, 383-391 (2004) Harue Mohri, Tomoko Amaya, Yoshiko Tsukahara, Yuko Miyake, Symptomatic Study of Micrographia, Higher Brain Function Research, Vol. 24, No. 1, pp. 1-10 (2004) Naoko Mikami, S-HTP method: Clinical and developmental approach using integrated HTP method, Seishinshobo (1995) Takeshi Nakayama, Hisao Maekawa, Satoshi Kami, Research on planning in character drawings of children with learning disabilities, Children's mind and nerve, Vol. 38, No. 2, pp. 91-105 (1998)

  For example, when the above-mentioned digital pen is used, the handwriting process of writing and drawing can be decomposed into strokes, and the position information and time information of points included in the strokes can be acquired and analyzed as time series data. Here, the stroke is equivalent to one stroke for kanji and one stroke for pictures. The digitized writing / drawing information can be expressed by approximating the actually written / drawing information within an error range that has no practical problem by connecting the points with broken lines.

  However, simply analyzing the stroke information alone cannot extract the feature values of “character” or “picture” and compare them with each other. For example, in order to evaluate dwarfism observed in Parkinson's disease patients, it is essential to distinguish and recognize individual “characters” and capture changes in their sizes. In addition, in the HTP test, it is not possible to evaluate the balance and positional relationship between each other unless you distinguish and recognize which part of the picture is a house, a tree, or a person. In addition, it is an important evaluation item that is drawn at which position on the drawing sheet and in what size.

  As mentioned above, focusing on information in units of “characters” and “drawings” drawn in everyday situations, quantifying and comparing and evaluating the features of “characters” and “pictures” No support means or device for judging the writer's mental, psychological, or physiological state has been provided.

  In order to solve this problem, the position information and time information of the points that make up “characters” and “pictures” drawn in daily life are time-series using input means such as a digital pen or tablet PC. Method and apparatus having means for distinguishing and cutting out each “character” or “picture” after being acquired as data, and means for extracting / displaying the feature amount of each distinguished “character” or “picture” Can be realized. The feature quantity extracted and displayed here provides a support device for comparing and evaluating each other.

  Examples of the feature amount of the cut out “character” or “picture” include the height, width, and area of the character and the picture. Furthermore, as in the case of the HTP test (House, tree, person test), the ratio of the height and the area between the three “pictures” of the house, the tree, and the person can be mentioned.

  According to the present invention, since comparison and evaluation results can be displayed quantitatively, doctors, counselors, etc. support high-accuracy determination of the writer's mental / psychological / physiological state, and further, the subject himself / herself, guardian, etc. It is possible to provide a determination result with objectivity to a third party.

  Mental health prevention and purchasing tendency by quantitatively grasping the writer's mental, psychological, and physiological state from handwritten writing and drawing data in everyday situations such as filling out questionnaires without doing anything special Analysis, learning efficiency improvement, and labor productivity improvement.

First, an outline of an embodiment of the present invention will be described.
The computer according to the present embodiment follows the flow 121 to 124 in FIG. 1B, the layout in the sheet, the position of the stroke entered in the sheet, the contact time of the pen stroke at the time of entry, and the interval between the strokes. Calculates and stores time, etc., distinguishes and recognizes each “character” or “picture” that is entered based on the stored information, and quantifies the character or picture features to compare and evaluate It is related with the assistance apparatus which judges a writer's mind, psychology, and physiological state by doing.

  The computer according to the present embodiment includes a digital pen connection device 101 that acquires stroke data from a digital pen, a recognition unit 103 that recognizes characters and pictures entered from stroke data, and an operation that compares and evaluates recognized characters and pictures. Part 102 is provided. The digital pen 111 includes a CPU, a memory, a communication interface, a camera 112, a battery, and a writing pressure sensor. In addition, the digital pen 111 includes a pen tip capable of writing characters, symbols, and the like on paper with ink or graphite.

FIG. 1A is a block diagram showing a logical configuration of a computer according to the embodiment of this invention.
The computer according to the present embodiment includes a digital pen connection device 101, a calculation unit 102, a recognition unit 103, a storage device 105, and an output device 106. Here, although the example stored in one computer 107 is shown, it is not necessarily limited to this, and it cannot be overemphasized that either may exist independently.

  The digital pen connection device 101 collects information from the digital pen 111 connected to the digital pen connection device 101. The calculation unit 102 includes a recognition unit 103, a feature amount extraction unit 104, a processor and a memory (not shown), and the processor performs various processes by executing a program stored in the memory.

  The recognizing unit 103 performs a character and picture cut-out process described on the paper. As an example of the clipping process, a stroke acquired by the digital pen 111 is recognized as a character using a character recognition database stored in the storage device 105. For this character recognition, an existing character recognition engine can be used. Similarly, when recognizing a picture, it is also possible to recognize a picture by executing image matching based on stroke information obtained from a digital pen using an image recognition database stored in the storage device 105. is there.

  Further, as an example of the cutout process, there is a method of cutting out characters or pictures by detecting spatial gaps between the projected images using projected images on one or a plurality of axes based on the stroke information. Further, a method of detecting a portion having a long stroke interval time by using stroke time information and cutting out characters or pictures may be employed. Further, the above two spatial and temporal cutout methods can be used together to increase the cutout accuracy. Furthermore, there is a method of detecting a difference in statistic such as a contour direction contribution degree or a contour local contour line from stroke information, and cutting out characters and pictures. The character / image recognition described above may be executed after executing these cutout processes.

  In the feature amount extraction unit 104, in addition to the coordinate information, paper ID, and writing time, information obtained from the digital pen as shown in FIGS. 4 and 5, the recognition result by the recognition unit 103, and the data in the storage device 105 are used. In addition, writing speed / acceleration, entry time in one character or picture unit, position on paper, height of character or picture, width of character or picture, ratio of character or picture in entry area on paper, character or The area of the picture is calculated as the feature quantity of the writing / drawing.

  Further, the standard value of the feature value described above is stored, and the degree of deviation from the standard value is calculated using the feature value database stored in the storage device 105.

  The storage device 105 is configured by a non-volatile storage device such as a magnetic disk device, and stores appearance frequency data 108, a standard time database 109, manuscript paper data 110, and the like as examples of a feature amount database.

  The appearance frequency data 108 includes information on strokes that are likely to appear during work. Specifically, it includes data on characters / pictures that are frequently or less frequently entered by the operator. It is possible to change the contents appropriately according to the scene to be used, and when there are many scenes to draw a picture, it is possible to store image data corresponding to the picture and write letters and numbers If there are many, letters and numbers can be stored accordingly. As a result, the processing speed required for the recognition process can be improved.

  The standard time database 109 includes information on the time required for writing characters and pictures. Specifically, for each character or picture, data of time required for entering a stroke and time required for moving the pen tip between strokes is included. The original paper data 110 includes information on the paper on which the user works.

  The storage device 105 may have a standard location database. This standard position database stores the standard positions of characters and pictures described on the entry form. In particular, in a scene where a description such as an HTP test (House, tree, person test) is performed, when the position, size, and range of drawing three “pictures” of a house, a tree, and a person can be an evaluation factor, One or a plurality of portions to be feature points are determined and stored in advance. For example, the coordinates of the center of each picture, or the coordinates of a portion having features such as the vertex portion of the tree, the boundary portion between the tree and the leaf, and the portion where the leaf is most overhanged. Information on these feature points can be used when calculating feature values to be described later. Further, the coordinates of these feature points may be stored in advance in association with the original paper data, or paper information is obtained by using the original paper data 110 from the information obtained by the input of the digital pen by the calculation unit 102 as appropriate. The standard arrangement of the picture may be specified from the paper size information obtained in this way, and may be associated with the paper surface position information such as a dot pattern corresponding to the specified position.

  The output device 106 is a display / printer or the like, and outputs a processing result by the computer. A storage device may be provided instead of the output device, and the processing result may be stored in the storage device.

Here, the principle of acquisition of stroke information by the digital pen 111 will be described.
FIG. 2 is an explanatory diagram of coordinate acquisition on the paper surface by the digital pen 111 according to the embodiment of this invention. The digital pen 111 is used together with the paper 201 on which dots 204 for position detection are printed.

  Here, the dot 204 will be described by enlarging a part 202 of the sheet 201. A plurality of small dots 204 are printed on the paper 201. The dots 204 are printed at positions shifted vertically and horizontally from the intersection (reference point) 203 of virtual grid lines. When characters and figures are entered on the paper with the digital pen 111, the entered characters and the like remain visible on the paper. When the digital pen 111 detects that the pen tip is in contact with the paper by the pen pressure sensor, the digital pen 111 photographs the dot 204 printed on the paper by the camera 112. For example, the digital pen 111 captures an area including 6 × 6 dots 204.

  The digital pen 111 calculates absolute coordinates where the dot pattern exists from the amount of deviation of the captured dot pattern. This absolute coordinate is a coordinate where the dot pattern exists in a vast plane area. This vast plane area is an entire area that can be arranged so that the same dot pattern does not overlap.

  The digital pen 111 transmits the calculated absolute coordinates to a server (not shown) via the digital pen connection device 101 and transmits it to the computer 107 via the server. Based on the absolute coordinates calculated by the digital pen 111, the server specifies the position of the paper surface (dot pattern ID) in the vast plane area and the coordinates (relative coordinates) on a certain paper surface, The dot pattern ID and relative coordinates are acquired from the dot pattern photographed by the digital pen 111.

  The digital pen 111 can acquire the movement of the pen tip by periodically acquiring information on the position where the pen tip is in contact (for example, every predetermined time). Then, the digital pen 111 transmits the absolute coordinates corresponding to the shot dot pattern, the shooting time of the dot pattern, and the pen ID to the computer 107.

  The computer 107 acquires relative coordinates from the position information server based on the absolute coordinates calculated by the digital pen 111. The server generates a locus (stroke information) of the pen tip from the acquired relative coordinates and the time when the dot pattern was captured. Here, an example via a server is shown, but a server is not always necessary, and the computer 107 may execute the above processing.

  Further, the digital pen 111 does not have to use a position information server in order to specify the dot pattern ID and the relative coordinates. For example, the paper may be specified by an identification number printed on paper, a barcode printed on paper, or an IC tag embedded in paper. Further, the position (relative coordinates) on the paper can be specified using a tablet. Note that either the specification of the paper using an IC tag or the like or the specification of the position on the paper using a tablet may be combined with the specification of absolute coordinates by the position information server. In this way, the document management system can reduce the process of specifying the dot pattern ID and relative coordinates.

  FIG. 3 shows an example of the stroke set 301 according to the embodiment of the present invention. The stroke set 301 indicates the characters “Tokyo” entered with the digital pen 111. In the present embodiment, as shown in the figure, the position of the stroke is determined with the upper left corner of the paper as the origin, the horizontal direction as the X axis, and the vertical direction as the Y axis. As described above, the stroke set is a set of a group of lines (strokes), and is specified based on the time when the lines are written and / or the positional relationship between the lines.

  FIG. 4 is a configuration diagram of the stroke set information 401 in which the stroke information according to the embodiment of the present invention is collected. FIG. 4 shows stroke set information of the stroke set 301 shown in FIG. The stroke set information 401 includes a stroke set ID 402, a pen ID 403, an entry start date and time 404, a corresponding rectangular coordinate 405, a stroke number 406, and stroke information 407.

  The stroke set ID 402 is a unique identifier of the stroke set. The pen ID 403 is a unique identifier of the digital pen 111 in which the stroke set is entered. The entry start date and time 404 is the date and time when entry of the stroke set is started. The corresponding rectangular area 405 is a rectangular area including the stroke set. The corresponding rectangular area 405 is the coordinates (relative coordinates) in the document in which the stroke set is written, and is represented by the upper left coordinates and the lower right coordinates of the rectangular areas. The stroke number 406 is the number of lines (strokes) included in the stroke set. The same number of stroke information 407 as the number of strokes 406 is recorded. The stroke information 407 includes a sample point number 407A and a serial number 407B. The number of sample points 407A is the number of relative coordinates acquired by the digital pen 111 and constituting the stroke. The serial number 407B is a unique identifier of relative coordinates that are acquired by the digital pen 111 and constitutes the stroke, and is a link to the stroke coordinate information 505 (FIG. 5).

  FIG. 5 is a configuration diagram of the stroke coordinate information 505 in the stroke information 23 according to the embodiment of this invention.

  The stroke coordinate information 505 includes a serial number 501, an X coordinate 502, a Y coordinate 503, and an acquisition time 504. The serial number 501 is a unique identifier of relative coordinates acquired by the digital pen 111. The X coordinate 502 is a relative coordinate in the X-axis direction and is expressed in units of millimeters, for example. The Y coordinate 503 is a relative coordinate in the Y-axis direction shown in FIG. 4 and is expressed in units of millimeters, for example. The acquisition time 504 represents the time when the relative coordinates are acquired by the digital pen 111. In the example shown in FIG. 5, an elapsed time from the entry start date and time 404 is recorded at the acquisition time 504.

  FIG. 6 is a flowchart showing the entire recognition processing by the computer according to the embodiment of this invention. A description will be given below mainly using the case of character recognition as an example.

  First, the digital pen connection device 101 acquires the acquired pen information (stroke information) from the digital pen 111 connected to the digital pen connection device 101 (601), and sends the acquired stroke information to the calculation unit 102. The computing unit 102 extracts a pen event (a pen down where the pen touches the paper surface and a pen up where the pen leaves the paper surface) from the acquired stroke information, divides the stroke information in the pen event, and identifies each stroke. (602). In addition, the calculation unit 102 creates an image entered by the user from the acquired stroke information, and sends the created image to the recognition unit 103. The recognition unit 103 performs image recognition or character recognition on the created image image (603).

  Thereafter, the calculation unit 102 compares the divided strokes with the strokes recognized as characters or pictures. Here, in the case of character recognition, a stroke that is not recognized as constituting a character or a picture is a result of a meaningless work, and therefore, it may be determined that it is a noise stroke. In this case, the determined noise stroke is removed from the divided strokes (604). By removing this noise stroke, it is possible to extract only the strokes entered by meaningful work, and to provide more accurate evaluation support.

  Further, it may be determined by referring to the appearance frequency data 108 whether the stroke is a noise stroke. For example, when other characters are entered in a scene where only characters / pictures are described, it is possible to determine that characters other than the target among recognized characters and pictures are noise.

  Thereafter, the calculation unit 102 specifies the start time and end time (entry start time and entry end time) of each stroke from which noise has been removed, and calculates the duration time and interval time of each stroke (605). Specifically, the stroke start time is obtained by subtracting the entry start time from the stroke entry end time. Further, the stroke interval time is obtained by subtracting the entry end time of the previous stroke from the stroke entry start time.

  Further, the calculation unit 102 specifies the start point and end point positions (X coordinate and Y coordinate) of each stroke from which noise has been removed, and obtains the movement time between the specified coordinates (606).

  Next, FIG. 7 shows an example of an analysis result of character information as an example obtained by the feature quantity extraction unit 104 and outputted to the output device. Analyzing time-series data consisting of position, time, and writing pressure information in writing, with time change of writing pressure 702, time change of writing speed 703, time change of writing acceleration 704, and sum of all stroke distances A total writing distance 705, an average writing speed 706, a character height and an area 707 are displayed.

  By using such a characteristic amount of writing, it is possible to estimate a change in writing speed and a mental / psychological / physiological state of the writer. For example, since the writing speed is gradually reduced, it is possible to estimate the writer's fatigue, reduced motivation, and the like.

  FIG. 8 shows an example in which character information (character string) 801 is analyzed and character height value 802 is calculated as a feature amount and graphed as an example obtained by the feature amount extraction unit 104 and output to the output device. Show. The horizontal axis of this graph is the position in the x-axis direction of a), and the horizontal axes of a) and b) correspond on the same scale. The character height was measured in the y-axis direction of a).

  Furthermore, the value of the height 803 of each character of the character string 801 is shown together in the table c). The character string 801 has two identical characters (for example, “day”). The same character height ratio 304 is summarized in the table d). The ratio value was calculated by (left character height) / (right character height). From the table in d), when focusing on the same character, it can be seen that the height of the right character is always smaller than the height of the left character. In other words, this indicates that the characters gradually become smaller as the writing proceeds from left to right, and is considered to be a typical example of dwarfism. It is possible to treat dwarfism quantitatively using the value of this ratio, and it can be applied to diagnosis support for Parkinson's disease and support for medication planning.

  In FIG. 9, as an example obtained by the feature quantity extraction unit 104 and output to the output device, the drawing information (picture) 901 is analyzed, and the values of the picture height 902 and the area 903 shown in b) are calculated as the feature quantity. An example is shown. Further, the absolute values of the height 902 and area 903 of the picture 901 and the ratio values based on the height and area of the person are summarized in Table c).

  The table in c) shows that the house is smaller than the size of the person. This is thought to be due to the fact that this picture contains a slightly perspective expression. In this example, no pathological factors such as mental instability are found.

  It is possible to handle the HTP test quantitatively using values of such ratios and quantitatively grasp the writer's mental, psychological, and physiological state. Furthermore, it can be applied to prevent mental health by monitoring changes in these data over time.

  FIG. 10 shows an example in which the writing information entered on the test question / answer sheet 1001 as shown in a) is collected using the digital pen 111. The written data is analyzed by the calculation unit 102, and when and in what order the characters are written on the answer sheet 1001 can be displayed in the form of "written time log 1002" as shown in b).

  From this written time log 1002, it is possible to determine the answer order of the questions and the thinking time. From this information, it becomes possible to analyze the respondent's thought process, such as which question the respondent is using, what kind of problem it takes to answer, and whether it is being checked or reviewed. I can grasp it. Therefore, it can be used to improve learning efficiency, formulate effective teaching methods, or improve labor productivity.

  FIG. 11 shows an example in which the letter information entered on the questionnaire sheet 1101 is collected by the digital pen 111. By analyzing the written data, a trajectory 1102 can be drawn as shown in the figure indicating from where and when in the questionnaire. In addition, as shown by the bar 1103 in the upper left, the time information displays the time zone of 15 seconds in shades.

  By analyzing the motivation, understanding, jealousy, self-confidence, etc. of the respondent from the information on the trajectory and the time, it becomes possible to grasp the object of interest and the importance of the respondent. Therefore, it can be applied to analysis of purchasing tendency. For example, there is a method of determining that there is a willingness to purchase if a threshold value is set for each index and the threshold is exceeded. Here, the number of responses is large (indicating high interest), the entry time is short (indicating that judgment is fast), the number of characters is large and the writing speed is fast (indicating that product knowledge is abundant), etc. A customer who meets one or more of the characteristics can be determined to have a high willingness to purchase, and the customer can quickly respond to explanations and treat customers carefully. Furthermore, if the written data of such a questionnaire is communicated by Bluetooth (registered trademark) wireless, etc., and analyzed in real time by a computing device serving as a base station, for example, a person with a high willingness to purchase is identified in the exhibition hall, It is possible to solicit on the spot. In addition, it is possible to grasp to some extent how much relevant knowledge you have about the questionnaire and whether you are interested in it. Can also be performed.

  The present invention quantitatively grasps the writer's mental, psychological, and physiological state from handwritten letters and drawing data in everyday situations, thereby preventing mental health, analyzing purchasing trends, improving learning efficiency, and labor production. It can be used to improve the property.

An example of the basic composition of the analysis system of the writing and drawing information in the present invention. Explanatory drawing of the coordinate acquisition on the paper surface by the digital pen of embodiment of this invention. An example of the stroke set of embodiment of this invention. The block diagram of the stroke set information of embodiment of this invention. The block diagram of the stroke coordinate information of embodiment of this invention. The flowchart which shows the whole process by the computer of embodiment of this invention. An example of the analysis result of written information. An example of calculating the change in character height from the text information of a character string. An example of calculating the ratio between the height and area of a picture from the drawing information of the picture. An example of a written time log generated from written data on test questions and answer sheets. An example of creating an entry track from a questionnaire form.

Explanation of symbols

101: digital pen connection device, 102: calculation unit, 103: recognition unit, 104: feature amount extraction unit, 105: storage device, 106: output device, 107: computer, 108: appearance frequency data, 109: standard time database, 110: Manuscript paper data, 111: Digital pen, 112: Camera, 121: Measuring device (input means) for writing information, 122: Recording device for writing data, 123: Arithmetic device for performing cut-out identification of characters and pictures 124: Calculation / display device that executes and displays feature amount extraction, 201: Paper on which dots are printed, 202: Part of paper 201, 203: Intersection of virtual grid lines, 204: Dots, 301: Stroke Set, 401: stroke set information, 402: stroke set ID, 403: pen ID, 404: entry start date and time, 405: corresponding rectangle Coordinates, 406: Number of strokes, 407: Stroke information, 407A: Number of sample points, 407B: Serial number, 501: Serial number, 502: X coordinate, 503: Y coordinate, 504: Acquisition time, 505: Stroke coordinate information, 601: Acquisition of pen information (stroke information), 602: division / specification of each stroke, 603: execution of image recognition or character recognition, 604: noise stroke removal, 605: calculation of duration and interval time of each stroke, 606: Calculation of movement time between specified coordinates, 701: Character, 702: Writing pressure data, 703: Writing speed data, 704: Writing acceleration data, 705: Total writing distance data, 706: Average writing Speed data, 707: Character height / area data, 801: Character string, 802: Character height data, 901: Picture, 9 2: picture of height, 903: the area of the picture, 1001: problem-answer sheet of the test, 1002: the writing time log, 1101: questionnaire, 1102: locus, 1103: bar that displays the time.

Claims (11)

Measuring handwritten handwriting information;
Recording the measured writing information;
Executing character and picture cutout identification based on the written information;
A mental state determination support method comprising a step of calculating a feature amount from stroke information in units of characters and pictures subjected to the cutout identification, and a step of displaying the feature amount. Measuring handwritten handwritten information described by a digital pen on a paper surface on which a dot pattern is formed; recording the measured handwritten information;
Performing character or picture cutout identification based on the written information;
Calculating a feature amount based on stroke information in units of the character or picture that has been subjected to the cutout identification;
Displaying the feature amount,
The feature amount includes at least one of a height and a width of the character or picture, a ratio of an area to be filled in on the paper, and an area. The mental state determination support method according to claim 1 or 2,
In the step of executing the cutout identification,
Based on the stroke information, using a projected image on one or a plurality of axes, a method for detecting a spatial gap between projected images, a method for detecting a portion having a long stroke interval time, a contour direction contribution degree, A mental state determination support method comprising a method for detecting a difference in statistics such as a contour local contour line as a clipping method. The mental state determination support method according to claim 1 or 2,
A mental state determination support method comprising a step of calculating a deviation degree between a standard value of the feature quantity stored in advance and the feature quantity. The mental state determination support method according to claim 4,
In the step of displaying the feature value, the degree of divergence is displayed together with the feature value. A digital pen connection device that acquires stroke information from a digital pen that can be written on a paper surface, a storage device that stores the stroke information, and a character that is recognized as a recognition unit that recognizes characters and pictures written from the stroke information In a mental state determination support device having a calculation unit having a feature amount extraction unit for comparing and evaluating pictures, and an output device,
Based on the stroke information and the recognition result, the feature amount extraction unit writes the height of the character or picture, the width of the character or picture, the ratio of the character or picture in the entry area on the paper, the area of the character or picture, etc. A mental state determination support device that calculates at least one of a character and a drawing as a feature amount. The mental state determination support apparatus according to claim 6,
The storage device stores the standard value of the feature amount in advance,
The feature quantity extraction unit calculates a divergence degree between the standard value and the feature quantity. The mental state determination support apparatus according to claim 7,
The mental state determination support apparatus, wherein the output unit outputs the degree of divergence together with the feature amount. The mental state determination support apparatus according to claim 8,
When multiple letters and pictures are entered on the paper,
The feature amount extraction unit includes:
A mental state determination support apparatus that calculates a degree of deviation from the standard value for each character or picture. The mental state determination support apparatus according to claim 8,
When multiple letters and pictures are entered on the paper,
The feature amount extraction unit includes:
A mental state determination support apparatus that compares feature amounts of characters and pictures with reference to predetermined characters and pictures and outputs the comparison result to the output means. The mental state determination support apparatus according to claim 7,
The mental state determination support apparatus characterized in that the feature amount includes any one of a writing speed / acceleration, a writing time for each character or picture, and a position on a paper surface.

Images