JP4088282B2 - Computer input method and apparatus - Google Patents

Description

  The present invention relates to a computer input method and apparatus using a pointing device that is used for pointing an arbitrary position on a screen displayed on a display of a computer or the like and executing processing.

  Conventionally, with the development of computer technology and the Internet, computers have come to be used in general households, and accordingly, man-machine interfaces have been developed for more comfortable use of computers. Currently, a keyboard and a pointing device are mainly used as means for inputting data to a computer, a communication device, or the like.

  Pointing devices are used for GUIs (Graphical User Interfaces) such as window systems, which are user interfaces designed using graphics components such as buttons and menus, rather than traditional character input interfaces. It is. Therefore, the pointing device is an input means used for pointing to an arbitrary place on the screen by the GUI. A mouse is a representative device of this pointing device.

  However, the operation of the mouse has a problem that it is unusable for a person with a handicapped person. For this reason, instead of the conventional mouse, there is a switch box provided with buttons in eight directions so that the operation by the button operation is close to that of the mouse. Recently, a computer input technique using the movement of the eyes of a user who uses a computer has been developed.

  As a method for detecting eye movement, a blinking switch has been developed in which infrared light is applied to the eyeball and the switch is turned on / off based on the difference in the amount of infrared reflection between when the eye is opened and when the eye is closed. There is also a line-of-sight input device that combines a cornea reflection method and a pupil image. In addition, in order to measure the line of sight of a person using a computer, there is a type in which a triangular infrared reflecting frame used for detecting movement of the head is attached to spectacles.

Furthermore, in patent document 1, the face of the user who uses a computer is imaged with a camera, the eyes | visual_axis of an eye is estimated from the imaged image, and the viewpoint position on the screen where an eye gazes is calculated | required. A method of operating a computer by displaying a cursor indicating a viewpoint on the screen, detecting an overlap with a menu displayed on the screen, and processing the overlapped menu has been proposed.
JP 2003-196017 A JP 2003-150942 A JP 2002-14765 A JP 11-110120 A Japanese Patent Laid-Open No. 10-143316

  Among the above conventional techniques, the method using a switch box instead of a mouse cannot be moved at all by a handicapped person.

  Further, in the line-of-sight input device that combines the corneal reflection method and the pupil image, it is difficult to obtain high accuracy because of the relative position correction between the corneal reflection point and the pupil center, and there is a problem of erroneous input. In addition, since the eyeball is irradiated with infrared rays, there is a problem with the eyes when used for a long time. Furthermore, there is also the trouble of fixing the head to the headrest. In addition, the computer operation support tool for persons with disabilities needs to prepare operation conditions and the like separately for each individual. In addition, in the method of attaching any device to the eyes or head, even a person with limited limbs can operate the computer, but it is bothersome to wear the device every time the computer is operated. There was something wrong with wearing it. In addition, there is a drawback that stress tends to accumulate when used for a long time.

  In Patent Document 1, these drawbacks are improved, the line-of-sight direction is determined without wearing anything on the eyes and head, and input data positioned in the line-of-sight direction is determined. However, a special menu screen used for operation is displayed on the computer screen, and the computer is operated by selecting the menu with a line of sight. In this method, since a general operation using a pointing device such as a mouse is different from an operation method, for example, general software cannot be operated in the same manner as when operating with a mouse or the like. For this reason, a menu screen corresponding to each software is required, which is not only poor in operability but also expensive.

  The present invention has been made in view of the above-described problems of the prior art, has operability close to a pointing device such as a mouse by manual operation, and can be easily operated even when a person with a limb or a hand cannot be operated. Another object of the present invention is to provide a computer input method and apparatus capable of operating a computer.

  The present invention relates to a camera capable of capturing a moving image, a storage device such as a computer that stores images captured by the camera in a predetermined cycle, and an eye operation by recognizing a position of a human face imaged by the camera. In the computer input method for performing computer input, the image information of the face of the person captured by the camera is scanned while comparing with a template that falls within the range of the iris size of the person's eyes, Detect the shade of the image with other parts of the face to detect the position of the eye, further determine the iris area at that eye position, determine the open / closed state of the eye based on the determined iris area, This is a computer input method for processing input.

The iris area of the eye is determined based on a predetermined range at the position of the eye, and the total of locations where the image density is equal to or greater than a predetermined threshold is equal to or greater than the predetermined threshold. The predetermined range for measuring the iris area of the eye is a rectangular range having a vertical and horizontal length enough to contain the iris of the eye. Then, in the rectangular range, the sum of the portions where the image darkness is equal to or greater than a predetermined threshold is determined to be an eye iris when it is equal to or greater than a predetermined value. The eye open / closed state determined by iris recognition when it is determined that the eye is closed when the sum of the areas where the image darkness is equal to or greater than a predetermined threshold within the rectangular range is equal to or less than a predetermined value. In the determination, when it is determined that the arbitrarily set fixed time is closed, a predetermined processing input is made by a click operation of the pointing device of the computer.

The pointing device of the computer has a function of judging the open / closed state of the eye based on the iris area recognized as the iris of the eye, and switching the valid / invalid state of the drag operation. If it is determined that the computer is stopped within a predetermined range, the drag operation of the pointing device of the computer is enabled, the holding by the pointing device is enabled, and the eye is open for a certain period of time and is stopped within the predetermined range. If it is determined that the pointing device has been dragged, the drag operation of the pointing device is invalidated and a drop operation is performed . Furthermore, the pointing device of the computer has a function of judging whether the eye is opened or closed based on the iris area recognized as the iris of the eye, and switching the scroll operation between valid and invalid. If it is determined that both eyes are closed for a certain time set in the above, the pointing device is enabled to enable screen scrolling, and the screen is scrolled up and down by moving the eyes up and down over a certain range. The scroll operation is turned off when it is determined that both eyes are closed for a predetermined time or more when the scroll operation is enabled .

The present invention also includes a camera capable of capturing a moving image, a storage device that stores images captured by the camera at predetermined intervals, and eye position recognition means for recognizing the positions of human eyes captured by the camera. In the computer input device, the eye position recognizing means is configured to store the image information of the human face captured by the camera within a range of the iris size of the human eye and the iris width template. scan in comparison with, is for detecting the position of the human eye to detect the density of an image of the iris portion and the eyebrows and other parts of the face of the eye, further, at the position of the eye, the eye The iris of the eye has a length and width that is sufficient to contain the iris of the image, and is larger than the template. Recognize that there is Of determining the open state, similarly to a total depth of locations equal to or greater than a predetermined threshold image, the blink detection unit that determines that the eyes are closed by equal to or less than a predetermined value, the blink detection unit Click determination means for determining whether or not the click operation is based on whether or not the closed state of the eyes detected by the step is a predetermined time or more, and a predetermined processing input of the computer is determined by the determination by the click determination means. Computer input device to perform.

Further, the iris area at the eye position is obtained, and based on the obtained iris area, it is determined whether or not the iris is stationary within a predetermined range for an arbitrarily set period of time. and stopping detecting means for the stationary state of the detected eye by the stopping detecting means, provided a drag determination means for determining whether a drug, the determination by the drag determination section to perform a drag operation by the computer It is .

  According to the computer input method and apparatus of the present invention, the operation of a computer has operability close to that of a conventional pointing device such as a manual mouse, even in a person with a handicapped person or in an environment where a hand cannot be used. The computer can be operated quickly and accurately.

  An embodiment of a computer input method and apparatus according to the present invention will be described below with reference to FIGS. The computer input method and apparatus of this embodiment uses a pointing device that can be operated with eyes instead of a general manual mouse as a pointing device for operating a computer.

  The configuration of this embodiment will be described with reference to FIG. In the configuration of the computer input device of this embodiment, a display 4, a keyboard 6, and a mouse 8 are connected to the computer 2, and an image input board (not shown) that captures camera video is attached to the computer 2. The image input board is connected to a camera 10 having a near-infrared high-sensitivity CCD and capable of capturing a moving image. The camera 10 is mounted on the upper portion of the display 4 by a support (not shown), and a red filter 14 is attached to the lens 12 of the camera 10. As the OS (Operating System) of the computer 2, a generally popular OS can be used. The eye-operated pointing device device driver includes two programs: an eye tracking program (hereinafter referred to as an eye trace program) and a mouse driver program for converting cursor display position information. The program is installed. In addition, a device driver for a general manual mouse is also installed.

  Here, a pointing device program by eye operation will be described. This eye operation program is developed and created using a commercially available programming language. For the inter-process communication between the two programs that make up this program, the existing named pipe method is used as shown in FIG. Used.

  Next, a pointing device program by eye operation will be described. First, the eye trace program will be described.

  As shown in the flowchart of FIG. 3, this eye trace program detects the position of the eye, tracks it, and detects blinking (closed eyes). First, the eye position detection method and tracking method of this program will be described. The face grayscale image acquired by the camera 10 is binarized by threshold processing, and template matching is performed on the binarized image using the one-eye portion template 16 shown in FIG. For this template matching, a residual successive detection method is used.

In the residual successive detection method, the difference from the template data is sequentially obtained for image information having the number of pixels of L × L pixels. As shown in FIG. 5, the matching evaluation value E (p, q) between the template t (x, y) and the partial input image f (x, y) on the captured image template is calculated by the following equation (1). In this method, the processing is aborted assuming that there is no correlation when a predetermined threshold value is exceeded, and the same position is calculated at the next position to find a matching position.
E (p, q) = Σ ^M−1 _{x = 0} Σ ^M−1 _{y = 0} | f (x + p, y + q) −t (x, y) | (1)
However, 0 ≦ p ≦ LM + 1, 0 ≦ q ≦ LM + 1

  Generally, as shown in FIG. 4, the diameter of the cornea of the adult eye 15 is about 11 mm (length: 9.3 to 11 mm, width: 10.6 to 12 mm). Since the iris 18 which is a black eye is under the cornea and its size substantially matches the cornea size, in this embodiment, the size of the iris 18 is set to 11 mm in diameter. The one-eye portion template 16 is formed by a cross-shaped straight line, has a horizontal width of 11 mm, a vertical width of 6 mm, and a line width of 1 pixel on the screen of the display 4. The number of pixels per mm is calculated based on the number of pixels by drawing a straight line of a certain length on the display 4. In this embodiment, 1 mm is 4.2 pixels, 11 mm is about 46 pixels, and 6 mm is about 25 pixels. By using such a cross-shaped template 16, even if the eye 15 is inclined, the circular iris 18 is not affected, and accurate matching with the iris 18 can be obtained.

  In this embodiment, template matching is performed by moving the one-eye portion template 16 by four pixels for speeding up. The matching degree of the horizontal width of the one-eye part template 16 is compared pixel by pixel, and the matching degree of the vertical width is compared to skipping one pixel, and when the matching degree of 90% or more of the horizontal width and 40% or more of the vertical width is satisfied simultaneously. Judge that it seems.

  Further, the template matching by the one-eye partial template 16 may cause some erroneous areas on the image to be detected by mistake. As a countermeasure against the mistake, as shown in FIG. 6, the features around the eyes 15 were used as check items. In this embodiment, the reference diameter of the iris 18 is 11 mm. Here, as shown in FIG. 6, the presence or absence of the eyebrows 19 on the top and bottom of the template 16 is used as a judgment material. For example, when the iris 18 is 11 mm, the check point a at the position of the eyebrow 19 is between 25 and 50 mm above the center of the template 16 in consideration of the difference due to individual differences in the human face. The check point b between 15 and 27 is also 15 to 27 mm, and the check point c below the eye 15 is also 15 to 22 mm, and is determined by comparing with the threshold value based on the binarized pixel values. As a result, it is determined from the binarized image data that there is no black portion immediately above the eye 15, the black portion by the eyebrow 19 is detected above, and the black portion is not below the eye 15. Can be a material. Further, as shown by a two-dot chain line in FIG. 6, even when the face image is enlarged, the eye position can be accurately determined by detecting the eyebrows 19 and the portion below the eyebrows 19 within a certain range. .

  The matching degree of these check items is 10% or more at the position of the eyebrow 19 in the portion above the eye 15, 20% or more at the position between the eyebrow 19 and the eye 15, and 20% at the position below the eye 15. As described above, when the matching of the check items is obtained, it is determined that the condition suitability of the eye 15 is satisfied. The final determination that the eye is 15 is made by template matching and feature check items around the eye 15. Thereby, even when the face image acquired by the camera 10 is enlarged or reduced, the position of the eye 15 can be accurately determined by detecting the eyebrows 19 and the portion below the eyebrows 19 within a certain range.

  Next, detection of blinking of eyes 15 (a state where eyes are closed) will be described. In the blink detection method of this embodiment, blinking is detected from a change in the area of the iris 18 that is a black eye. As shown in FIG. 3, in the blink detection process, the eye 15 is detected by performing template matching using the template 16 of the one-eye portion from the grayscale image acquired by the camera 10 in the above-described steps, and then the iris is detected. 18 areas are measured. When the area measurement result of the iris 18 is equal to or less than the threshold value, it is determined that the eye 15 is closed.

  The area of the iris 18 is measured by scanning the binary image of the eye 15 with the measurement range s shown in FIG. 7 as a unit. Then, when the degree of matching between the width of the iris 18 and the one-eye partial template 16 satisfies 90% or more, it is determined that the eye 18 is visible, and it is considered that the width of the iris 18 at the time of eye opening coincides with the width of the template 16. Further, when the eye is closed, a long black portion is formed in the horizontal direction by the eyelash 20, and therefore the horizontal measurement range s 1 of the area of the iris 18 is set to the horizontal width of the template 16 in order to prevent a large area from being measured. The vertical measurement range s2 is a vertical direction from the center of the template 16 in consideration of the fact that the match point deviates from the center of the iris 18 because there are many matching points of the one-eye portion template in the iris 18. The range was 50 pixels.

  In the iris area measurement procedure, after the eye 15 is recognized, the area of the iris 18 is measured. As shown in FIG. 8A, the template 16 moves upward from the center position within the vertical measurement range s2 in FIG. 7 until the pixel density value is outside the threshold range of the iris 18, and the movement distance thereof. Calculate Next, the distance is calculated in the same manner in the downward direction, and the sum of the vertical movement distances is set as the vertical width of the iris 18. Then, the same operation is repeated up to the left end of the template 16 within the measurement range s1, and the vertical width is totaled. Similarly, the vertical width is totaled up to the right end of the template 16. This measurement is performed in the same manner when the eyes are closed. When the eye is closed, as shown in FIG. 8B, a black portion due to the eyelashes 20 appears long in the horizontal direction. However, the horizontal measurement range s <b> 1 is the horizontal width of the template 16, and the darker portion of the iris 18 than the threshold is the vertical width of the eyelash 20 in the vertical direction. Therefore, the area of the iris 18 at the time of eye opening becomes a larger value within the measurement range s than when the eyelash 20 at the time of eye closing is measured.

  As a result, if the total of the locations where the density is greater than or equal to the threshold (the above moving distance) within the measurement range s in FIG. Further, if this sum is equal to or less than the threshold value, it is assumed that the eyes are closed and the eyes are closed.

  Next, a mouse driver program that executes a mouse function based on eye tracking information of the eye tracing program will be described. This program performs a mouse function by controlling the cursor on the screen based on the eye position coordinates detected by the eye trace program.

  The moving method for moving the cursor of the mouse driver program includes an absolute coordinate mode and a vector mode. In addition, functions common to each mode include functions such as navigation window display, eye tracking effective range setting, and desktop effective range setting.

  Here, each function will be described. The navigation window 21 is displayed at any one of the four corners of the screen, for example, as displayed on the display 4 in FIG. The navigation window 21 displays the current state of the eye operation program. Display content includes the size of the face image captured, cursor control enabled / disabled, cursor movement mode, tracking eye selection, currently detected eye position, cursor moving effective range, etc. It is. The eye tracking effective range is set by displaying the eye tracking effective range setting screen 22 shown in FIG. 9, displaying the acquired image from the camera 10 on the screen, and setting the eye tracking detection range for the image. Set. The maximum detection range is, for example, 640 × 480 dots. Further, whether or not the mouse can be moved outside the effective range of the eye for moving the cursor can be selected. If the mouse cannot be moved outward, it will not move outside the desktop range shown in FIG. Note that the desktop effective range setting screen 24 in FIG. 10 can be displayed by checking the desktop range check box on the screen 22 shown in FIG. 9 and clicking a button.

Next, a method for moving the cursor will be described. First, the absolute coordinate mode will be described. The image size input from the camera 10 is fixed, and the eye position when the computer user moves his / her face is limited to a certain range of the camera image. Therefore, as shown in FIG. 11, the absolute coordinate mode is a method in which the eye position range for the camera image is selected as a rectangle, and the eye position in the rectangle is directly relative to the cursor position in the desktop. At this time, the equations for calculating the cursor position in the desktop range from the eye position are as shown in equations (2) and (3).
cx = (x−bx) / w × w _d + bx _d (2)
cy = (y−by) / h × w _d + by _d (3)

  On the other hand, in the vector mode, as shown in FIG. 12, when the computer user views the screen, the eye position is at a fixed position. In this method, the position is set as the center point, the direction from the center to the eye position is converted into the moving direction of the mouse cursor, and the distance from the center point to the eye position is converted into the cursor moving speed.

  Next, the function of the pointing device by eye operation of this embodiment will be described. This pointing device function is virtually equipped with a manual mouse function, and includes a click function, a double-click function, a drag function, and a scroll function.

  First, the click function and the double click function will be described. The determination of click and double click is performed according to the algorithm represented by the flowchart shown in FIG. In this algorithm, the role of the first button of the mouse designed for right-handedness is assigned to the movement of the left eye, and the role of the second button is assigned to the right eye. The determination corresponding to the click of the first button of the mouse is determined to have been clicked by closing the left eye for a certain period of time. Similarly, the determination corresponding to the click of the second button of the mouse is determined to have been clicked by closing the right eye for a certain period of time. If it is determined that it has been clicked, for example, the mark “+” indicating the eye position displayed in the navigation window is changed to red to indicate that it has been recognized. The determination of the click is made based on the time when the eyes are closed as shown in FIGS. 13 and 14, that is, the iris area by the above-described blink detection process. The click is determined by detecting the eye from the image (frame) by the camera 10 as described above, measuring the iris area, and determining based on the time when the iris area is equal to or less than a predetermined threshold. In the determination, the time during which the iris area is equal to or less than a predetermined threshold is in the range of 0.5 s to 1.0 s. For example, at the point A in FIG. 13, since the area of the iris 18 becomes equal to or greater than the threshold value, that is, the blinking time is less than 0.5 seconds (s), it is determined as normal blinking. To do. On the other hand, at the point B, it is determined to be a click because the time when the eyes are closed is 0.5 s to 1.0 s.

  The double click is determined as a double click by performing the above click operation twice. If the interval between the first click and the second click is within a certain period of time, it is determined as a double click. If it is more than the time, it is determined that the single click is performed twice. When the double click is determined, for example, the “+” mark indicating the eye position of the navigation window is made yellow. As shown in FIG. 15, the condition for determining a double click is that the iris area is equal to or less than a threshold value for a click determination time of 0.5 s to 1.0 s, When the condition to be determined is satisfied, it is determined as a double click. In this flow, it is determined that the click is made at the point A in FIG. 15 and another click is made at the point B. However, as shown in FIG. Depending on whether or not, a single click or double click process is executed.

  Next, the drag function will be described. FIG. 16 is a flowchart showing the drag determination algorithm. By keeping the eyes within a certain range for a certain period of time in an open state, the drag mode is turned on and it is determined that dragging is possible. When the eyes are closed so as not to be confused with a click, it is not determined that the drag mode has been turned on even if the eyes remain in a certain range for a certain time. At this time, for example, the color of the “+” mark indicating the eye position of the navigation window is changed to display that the drag is recognized. FIG. 17 shows the eye position of the drag determination. When an eye is detected by template matching, position information where the eye is stopped at the time of detection is acquired, and the position is set as a reference point 32. From this point, it is determined whether the vehicle is stopped within a certain range s vertically and horizontally. In this case, the size of the fixed range s is set to 23 pixels vertically and horizontally by default. Further, if the eye position 34 is stopped within the range s, it is determined whether or not the eye position is stopped for a certain time. In this case, the initial setting for a fixed time is 1.0 s. Then, as shown in FIG. 17 (a), if it can be determined that the position 34 for a certain period of time is stopped in the certain range s, the drag mode is turned ON. In order to end the drag operation and drop, when the drag mode is ON, it is dropped again when it is determined that the vehicle stays in a certain range for a certain time, and the drag mode is turned OFF and the drag ends. Further, as shown in FIG. 17B, when the position 34 at the fixed time is not stopped in the fixed range s, the process returns to the acquisition of the image (frame) at the next eye position as shown in the flowchart of FIG.

Next, the scroll function will be described. FIGS. 18 and 19 are flowcharts showing an algorithm for scroll determination. As shown in this flowchart, the iris area is measured for each of the left and right eyes, and when it is determined that both eyes are simultaneously closed for a predetermined time or more, the scroll mode is turned on. In order to turn off the scroll mode, the scroll mode is turned off by closing both eyes for a certain period of time as in the case of turning on . At this time, in the navigation window, the display state is the same as that in the vector mode, and the characters “SCROLL” are displayed. Further, an indicator representing scroll sensitivity is displayed in the navigation window. Scrolling is only possible to scroll up and down, with the circle representing the position range of the navigation window as the center, scrolling up when you move your eyes above the circle and scrolling down when you move down.

  According to the computer input method and apparatus of this embodiment, since an eye is used as a pointing device, a normal manual mouse can be used by using this eye operation mouse even in a situation where a handicapped person or a hand cannot be used. The computer can be operated close to when it is used. Further, this computer input method is operated by the program of the above-described embodiment installed in the computer, and it is not necessary to attach a special device to the eyes or head, and can be widely used in commercially available computers. Is. In addition, a general manual mouse can be used in a connected state. Usually, a manual mouse is used as a pointing device, and this eye operation mouse can be used in situations where a handicapped person or hand cannot be used. it can. For this reason, since it can be used also as a computer, it is not necessary to dedicate a computer and it is economical.

  Note that the computer input method of the present invention is not limited to the above embodiment, and which of the left and right eyes is assigned to the first button of the mouse can be selected by a key operation on the keyboard. The determination time for the click determination and the double click determination can be set as appropriate. Furthermore, the number of pixels representing the range used for drag determination and the threshold time of the stop time can be set as appropriate. In addition, a click by blinking can be replaced with a click by stopping eyes. In addition, since the algorithm of the computer input method of the present invention is portable, the OS used by the computer can be selected as appropriate.

It is the schematic which shows the apparatus structure of the computer input device of one Embodiment of this invention. It is the schematic which shows the data delivery of this embodiment. It is a flowchart which shows the blink detection of this embodiment. It is the schematic which shows the one-eye part template of this embodiment. It is the schematic which shows the relationship between the image of a residual successive detection method of this embodiment, and a template. It is the schematic which shows the position of the check point of the peripheral region feature of the eye of this embodiment. It is the schematic which shows the measurement range of the area of the iris of this embodiment. It is the schematic which shows the area measurement method of the iris of this embodiment. It is an effective range setting screen of eye tracking of a display display of this embodiment. It is a desktop effective range setting screen of the display display of this embodiment. It is a conceptual diagram which shows the conversion conditions of the absolute coordinate mode of the display display of this embodiment. It is a conceptual diagram which shows the conversion conditions of the vector mode of the display display of this embodiment. It is a flowchart of the click determination of this embodiment. It is a conceptual explanatory drawing which shows the determination time of the click determination of this embodiment. It is a conceptual explanatory drawing which shows the determination time of the double click determination of this embodiment. It is a flowchart of the drag determination of this embodiment. It is explanatory drawing which shows the position of the eye of the drag determination of this embodiment. It is a flowchart of the scroll determination of this embodiment. FIG. 18 is a flowchart subsequent to the scroll determination flowchart of FIG. 17. FIG.

Explanation of symbols

2 Computer 4 Display 6 Keyboard 8 Mouse 10 Camera 12 Lens 14 Filter 15 Eye 16 Template 18 Iris

Claims (5)

A camera that can capture moving images, a storage device that stores images captured by the camera at predetermined intervals, and a computer that recognizes the position of the eyes of a person's face captured by the camera and performs computer input by operating the eyes In the input method, the image information of the face of the person captured by the camera is scanned while comparing with the iris width template having a size that falls within the range of the iris size of the human eye, Detecting the density of the image of the iris part and the eyebrows and other parts of the face to detect the position of the eyes , and further having a vertical and horizontal length at which the iris of the eye enters at the eye position and from the template range rectangle also large, the total density of more locations a predetermined threshold image, determines that the open state of the eye recognizes that it is the iris of an eye by the predetermined value or more, likewise the In the range of the rectangle It is determined that the eyes are closed when the sum of the areas where the image darkness is equal to or greater than a predetermined threshold is equal to or less than a predetermined value, and is arbitrarily set in the eye open / closed state determination obtained by iris recognition of the eyes. A computer input method, characterized in that when it is determined that a predetermined time is closed, a predetermined processing input is made by a click operation of a pointing device of the computer. The pointing device of the computer has a function of judging the open / closed state of the eye based on the iris area recognized as the iris of the eye, and switching the valid / invalid state of the drag operation. If it is determined that the computer is stopped within a predetermined range, the drag operation of the pointing device of the computer is enabled, the holding by the pointing device is enabled, and the eye is open for a certain period of time and is stopped within the predetermined range. The computer input method according to claim 1 , wherein when it is determined that the dragging operation of the pointing device is invalidated, the drop operation is performed . The pointing device of the computer has a function of judging whether the eye is opened or closed based on the iris area recognized as the iris of the eye and switching the scroll operation between valid and invalid, and is arbitrarily set when the scroll operation is invalid If the eyes certain period of time that is determined to be closed, the enable pointing device, to allow scrolling operation, then scrolling vertically the screen by moving the eye over a certain range in the vertical 2. The computer input method according to claim 1 , wherein when the scroll operation is valid, if it is determined that the eyes are closed for a predetermined time or more, the scroll operation is turned off . In a computer input device having a camera capable of capturing a moving image, a storage device for storing images captured by the camera in a predetermined cycle, and eye position recognition means for recognizing the position of the eyes of a person captured by the camera The eye position recognizing means scans the human face image information captured by the camera while comparing with the iris width template having a size that falls within the range of the iris size of the human eye. And detecting the human eye position by detecting the density of the image of the iris part of the eye and the eyebrows and other parts of the face, and the degree of the eye iris entering at the eye position. When the sum of the areas where the image darkness is equal to or greater than a predetermined threshold is greater than or equal to a predetermined value within a rectangular range that is greater than the template in length and width, the image is recognized as an iris of the eye. Judgment with eyes open , Similarly, the total density of points equal to or greater than a predetermined threshold image, the blink detection unit that determines that the eyes are closed by equal to or less than a predetermined value, has been eye detected by the blink detection unit Click determination means for determining whether or not the click operation is performed based on whether or not the closed state is longer than a predetermined time, and performing predetermined processing input of the computer based on the determination by the click determination means. Computer input device. A stop detecting means for detecting a stop state of the eye by determining whether the iris recognized at the position of the eye is stopped within a predetermined range for an arbitrarily set time, and an eye detected by the stop detecting means Drag determining means for determining whether the drag operation is enabled or disabled in the stationary state, and when the drag operation is enabled by the determination by the drag determination means , the drag operation by the computer is performed The computer input device according to claim 4.

Images