JP2000028322A - Device for detecting high speed moving object position - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a high speed moving object position detecting device for detecting the position of an object moving at a high speed using an artificial retina chip having a pattern matching function available as a human interface system. SOLUTION: This position detecting device is provided with a first inputting means 36a on which an artificial retina chip having a pattern matching function for detecting the X/Y coordinates of a moving object is loaded, second inputting means 36b on which an artificial retina chip having a pattern matching function for detecting the Z coordinate of the moving object at the same time is loaded, and control processing means (micro-computer 39) for calculating the orbit of the moving object from data obtained by the inputting means 36a and 36b.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-speed moving object position using an artificial retinal chip having a pattern matching function, which can be used as an intelligent and flexible human interface system for allowing a computer and a human to interact with each other. The present invention relates to a detection device.

[0002]

2. Description of the Related Art A hand gesture recognition system will be described as an example of a conventional human interface system. FIG. 8 shows, for example, IEICE Transactions, D-II, Vol. J73-D-II, No. 12, PP-
It is a block diagram of the hand gesture recognition system published in 1985-1992 (1990). 8, reference numeral 1 denotes a hand shape input device (a commercially available data grab TM “Dat
a Glove TM "). Data Grab TM is a product of the United States VPL, which measures the bending information of a finger by the amount of light in an optical fiber, and uses the second and third joints of the thumb and the first and second fingers of the other finger. The bending angle corrected for individual differences is measured for 10 joints of two joints.A three-dimensional digitizer (3S) manufactured by MacDonnell Douglas, USA using a magnetic generator and a magnetic sensor.
By combining PACE TRACKER TM), hand orientation and hand 3
Dimensional position can also be measured.

[0003] 2 is a corrector, 3 is a voice recognition device including a microphone and the like, 4 is a voice synthesis device including a speaker and the like, 5 is an image display device, 6 is a work station, and 7 is a device for recognizing hand shape and movement. 8 is an analysis program for analyzing speech recognition results, and 9 is for integrating recognition results of hand shape / movement and speech analysis results, and outputting designated contents in the form of speech or images. Conversion program, 1
0 is an image database.

Next, the operation will be described. The hand gesture is input to the workstation 6 by the hand shape input device 1 and the personal difference corrector 2, and is recognized by the recognition program 7 for recognizing the shape and movement of the hand. After the voice information is recognized by the voice recognition device 3,
It is sent to an analysis program 8 for analyzing the speech recognition result of the workstation 6.

[0005] The recognition result of the hand gesture and the voice is
The result of hand shape / motion recognition and the result of voice recognition are integrated and sent to an external voice synthesizer 4 or image display 5 by a conversion program 9 for outputting the specified contents in the form of voice or image. .

[0006]

Since the conventional human interface system is configured as described above, a hand shape input device such as a glove (1 in FIG. 8) is required for hand gesture recognition. Yes, it was a very complicated and large system, requiring a workstation for information processing. Therefore, there is a problem that it is difficult to use it as a general-purpose human interface system.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and has a wide application range, operates at high speed, and has a more compact human body by detecting and recognizing hand gestures and body actions by an image sensor. An object of the present invention is to provide an interface and a high-speed moving object position detecting device using the same.

[0008]

SUMMARY OF THE INVENTION In view of the above-mentioned object, the present invention relates to a first input unit equipped with an artificial retinal chip having a pattern matching function for detecting the XY coordinates of a moving object, Calculating a trajectory of the moving object from data obtained by a second input unit equipped with an artificial retinal chip having a pattern matching function for detecting the Z coordinate of the moving object and data obtained by the first and second input units; And a control processing means.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below according to each embodiment. Embodiment 1 FIG. FIG. 1 shows a configuration of a human interface system according to an embodiment of the present invention. In the figure, 11 is a person, 12 is an image sensor composed of an artificial retinal chip and a CCD, 13 is a voice input device composed of a microphone or the like, 14 is a microcomputer (hereinafter referred to as a microcomputer), 15 is a human interface module, 16 is gesture recognition and voice. A recognition algorithm (software) 17 is a controlled device including a multimedia device including a personal computer, a home electric appliance (television, air conditioner, audio, etc.), a game machine, a care device, and the like.

The image sensor 12 constitutes a gesture input means for inputting a human gesture as image information, and the voice input device 13 constitutes a voice input means for inputting a human voice as voice information. Algorithm 16 constitutes an algorithm storage means,
The microcomputer 14 constitutes information recognition processing means.

Next, the operation will be described. The person 11 expresses his / her intention by hand gesture or body action. As the image sensor 12, an artificial retina chip or a CCD is used. The artificial retinal chip (registered trademark of Mitsubishi Electric Corporation) has a structure shown in FIG. Here, reference numeral 18 denotes an artificial retinal chip substrate, 19 denotes a random access scanner, 20 denotes a multiplexer, 21 denotes a pixel core circuit, 22 denotes a sensitivity control signal, 23 denotes an input image, and 24 denotes an output image.

The pixel core circuit 21 can change the light detection sensitivity from positive to negative, and the random access scanner 19 generates a control signal 22 for changing the light detection sensitivity of the pixel core circuit 21. . For example, the sensitivity of each of the first and second rows of the pixel core circuit 21 is set to 1,
Assuming that the sensitivity of the third row to the last row is 0 (current does not flow even when light is applied), a difference signal between the irradiated input images 23 is obtained in the first row and the second row. Accordingly, by scanning the pair (1, -1) of the sensitivity control signals 22 from the first row to the last row, an edge detection image of the input image 23 is obtained as the output image 24.

If the sensitivity control signal 22 is 1, 0, 0,..., The input image 23 can be detected as it is (CCD function). If all the sensitivity control signals 22 are set to 1, a projection signal of the input image 23 is obtained in the output image 24. The artificial retinal chip also has a feature that image detection and processing can be performed at high speed (frame rate: 1 kHz or more). The signal preprocessed by the retinal prosthesis chip is processed by the microcomputer 14 to perform recognition processing of a hand gesture and a body action of a person. Note that the microcomputer 14 also controls the contents of image processing in the above-described artificial retinal chip.

The voice is a voice input device 1 such as a microphone.
3 and is subjected to spectral decomposition, phonological decomposition and the like by a signal processing device such as the microcomputer 14 and recognized.

The results of the hand gesture, body action recognition, and voice recognition are integrated into a personal computer,
A control signal is input as a control signal to a controlled device 17 such as a home appliance (television, air conditioner, audio, etc.), a game machine, a nursing device, and the like, and the controlled device 17 can be operated according to the intention of the person 11. These recognition algorithms 16
Is mounted in association with, for example, the microcomputer 14 which is an information processing device.

In the above-described embodiment, an artificial retinal chip having an image pre-processing function (edge detection, projection detection, etc.) is used as an image sensor, but a CCD without an image processing function may be used. It is. When a CCD is used, both image preprocessing and signal processing must be performed by an information processing device such as the microcomputer 14.

Further, in the present invention, it is needless to say that the voice input device 13 for inputting human voice is not indispensable and may be omitted.

Embodiment 2 FIG. FIG. 3 shows a pattern of information input by hand gesture, which is a feature of one embodiment. In FIG. 3, 15a is a human interface module equipped with an artificial retinal chip and a microcomputer, 17 is a controlled device such as a personal computer,
11a is a user. As shown in FIG.
The hand gesture of “a” is recognized by the compact human interface module 15a, and the controlled device 17 such as a personal computer, a home appliance, a game machine, or a nursing device can be controlled based on the hand gesture.

FIGS. 4A to 4D show examples of hand gesture recognition. In the figure, reference numeral 28 denotes an imaging area (monitor screen) of the retinal prosthesis chip, 29 denotes a hand, 30 denotes an approximate hand shape as a rectangle, and 31 denotes a center of gravity. As shown in FIG. 4, the hand gesture approximates the shape of the hand as a rectangle, and its center of gravity coordinates (X1, Y1), size (d1 × d2), and inclination angle
(θ). Thus, various hand gesture recognitions can be performed with a simple and small amount of information.

Embodiment 3 FIG. 5 shows a feature of another embodiment, in which a user's intention (switch on, off, etc.) is recognized by recognizing a human body action. An example of a body action is shown in FIG.
To (f). In the figure, 32a to 32f indicate the position of the body action, particularly the arm position (up and down, left and right, diagonal, horizontal, etc.). In addition to the above, it is also possible to recognize body actions such as right and left inclination, squatting, jumping and the like of the body. The body action can be recognized by comparing the basic body action registered in advance with the input body action.

Embodiment 4 FIG. FIG. 6 shows an example of application to a game machine using a hand gesture or a body action as an input, which is a feature of another embodiment. In the figure, 15b is a human interface module, 33 is a game screen, 11b is a player, and 35 is a character in the game. As shown, player 1
The character 35 of the game operates according to the body action 1b. FIG. 6 shows an example applied to a flying game. According to the operation of the player 11b,
Since the character 35 flies in the sky, the player 11b can feel as if he / she is flying.

As described above, using the hand gesture and the body action as input, a game machine that does not require a joystick, a game machine in which characters perform the same operation as a player, or a keyboard, a mouse, a switch, a dial, a remote controller, and the like are used. Instead, a human interface system that can control a computer, a home appliance, and a nursing device can be realized.

Embodiment 5 Next, FIG. 7 shows a configuration of a high-speed moving object position detecting device using a human interface system according to another embodiment. In FIG. 7, the position of an object moving at high speed is detected. In the figure, reference numerals 36a and 36b denote cameras mounted with an artificial retinal chip, 37a to 37e denote planes on which moving objects are displayed at fixed time intervals, 38a to 38e denote moving objects, 3
Reference numeral 9 denotes a microcomputer for calculating and processing the trajectory of the moving object based on the input from the cameras 36a and 36b equipped with the artificial retina chip.

The artificial retina chip mounted camera 36a,
36b constitutes first and second input means, and the microcomputer 39 constitutes a control processing means.

Next, the operation will be described. The artificial retinal chip has a pattern matching function. That is,
Since an image of a specific pattern can be extracted from the input image, only the moving object can be detected by pattern matching the shape of the moving object (a circle for a ball). Therefore, the position (coordinates) can be detected immediately. The camera 36a with an artificial retinal chip is for detecting the X-Y coordinates of a moving object.
b is for detecting the Z coordinate of the moving object.

That is, at time t1, the XY coordinates of the moving object 38a on the plane 37a are represented by the camera 36a with the artificial retinal chip, and the Z coordinates are represented by the camera 36 with the artificial retinal chip.
b, the moving object 3 at time t1
The coordinates of 8a are known. Similarly, times t2, t3,...
, The coordinates of the moving objects 38b, 38c,... Can be respectively detected. Since the artificial retina chip can perform high-speed image detection and processing (frame rate: 1 kHz), it can detect the coordinates of a moving object every 1 msec.

If the moving object is, for example, a thrown ball, its trajectory is a parabola (quadratic curve y = aX2 + bX +
In order to draw c), if the X-Y coordinates and the Z coordinates (indicating the distance) of at least three different times are known, the locus can be calculated by the microcomputer 39 by calculation. If you know the trajectory, you can also know where the ball fell.

[0028]

As described above, according to the present invention, the first input means equipped with the artificial retinal chip having a pattern matching function for detecting the XY coordinates of the moving object, and the Z coordinate detection of the moving object at the same time The trajectory of the moving object is calculated from the data obtained by the second input means equipped with the artificial retinal chip having a pattern matching function for performing the moving object position, so that the position of the moving object can be detected at high speed. Effects such as the provision of a detection device can be obtained.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a human interface system according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of an artificial retinal chip.

FIG. 3 is a diagram showing a pattern of information input by hand gesture, which is a feature of one embodiment of the present invention.

FIGS. 4A to 4D are diagrams showing examples of hand gesture recognition.

FIG. 5 is a diagram illustrating an example of a body action when recognizing a human body action which is a feature of one embodiment of the present invention;

FIG. 6 is a diagram showing an example of an application to a game machine using a hand gesture or a body action as an input, which is a feature of one embodiment of the present invention.

FIG. 7 is a diagram showing a configuration of a high-speed moving object position detecting device using a human interface system according to another embodiment of the present invention.

FIG. 8 is a configuration diagram of a conventional hand gesture recognition system.

[Explanation of symbols]

11 person, 11a user, 11b player, 1
2 image sensor, 13 voice input device, 14, 39 microcomputer, 15, 15a, 15b human interface module, 16 gesture / voice recognition algorithm, 17 controlled device, 18 artificial retinal chip substrate,
33 game screen, 35 characters, 36a, 36b
Camera with artificial retinal chip.

 ──────────────────────────────────────────────────の Continuing on the front page (72) Kenichi Tanaka, 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsui Electric Co., Ltd. (72) Atsushi Ota 2-3-2, Marunouchi 3-chome, Chiyoda-ku, Tokyo Rishi Electric Co., Ltd. (72) Inventor Nobumi Komori 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Sanrishi Electric Co., Ltd. (72) Inventor William T. Freeman Alert, Newton, Mass., USA Download 174

Claims (1)

[Claims] A first input unit having an artificial retinal chip having a pattern matching function for detecting XY coordinates of a moving object; and a pattern matching function for detecting Z coordinates of the moving object at the same time. A high-speed moving object position detecting device comprising: a second input unit equipped with an artificial retinal chip; and a control processing unit for calculating a trajectory of the moving object from data obtained by the first and second input units.