JP2005209028A - Portable terminal and collision danger determination report method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect at an early stage and with good accuracy an obstacle having possibility of collision. <P>SOLUTION: A contour extract section 104 extracts the contour of object picture data of entire picture data from a camera 103 to generate object contour data. A feature detection section 105 detects the amount of a feature of the object contour data. An obstacle recognition section 107 compares the amount of the feature with obstacle feature data, recognizes the obstacle, and generates obstacle recognition data. A movement detection prediction section 108 detects the movement of the object picture data based on a time change in the amount of the feature, predicts the movement, and generates movement prediction data. A danger determination section 109 determines collision danger based on the obstacle recognition data and the movement prediction data, and generates the degree of collision danger. A danger report section 112 reports the collision danger when the degree of the collision danger becomes or exceeds a threshold. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a mobile terminal device having a function of determining and notifying a collision risk and a collision risk determination notification method.

  With the widespread use of mobile phone terminal devices, there are situations where people walk while operating the mobile phone terminal devices on roads or pedestrian paths. In these cases, since the mobile phone terminal device is used as an image information terminal device, an operation of gazing at the mobile phone terminal device by an e-mail operation, a browsing operation or a game operation when connected to the Internet is often performed. ing. For this reason, on the road or pedestrian passage, the discovery of obstacles such as people approaching from the traveling direction, bicycles or power poles is delayed, and there is a risk that the user of the mobile phone terminal apparatus collides with the obstacles.

A cellular phone terminal device or a portable information terminal device capable of detecting an obstacle and notifying its presence is described in Patent Document 1. This conventional cellular phone terminal device is provided with any one of an infrared sensor, an ultrasonic wave, and a camera as an obstacle detection means, and the detection area of the detection means is set to 1 to 2 m forward and the width is 45 degrees. By setting to a certain level, if an object enters the detection area, it is detected as an obstacle.
JP 2003-240850 A

  However, in the conventional mobile phone terminal device, since an obstacle is detected when an object enters the detection area of the detection means, an obstacle such as a bicycle or a mobile phone moving at high speed when the detection area is small In a situation where the user of the terminal device is walking at high speed, by the time it takes for the user to recognize the obstacle detection and alert information from the mobile phone terminal device, the user May collide with obstacles.

  Also, in the conventional mobile phone terminal device, when the detection area is widened so as to detect an obstacle from a distance, the detection of an obstacle detection is frequent because it detects even an object with no possibility of collision. There is a problem of doing.

  Furthermore, in the conventional mobile phone terminal device, when there is no risk of collision with a pedestrian moving in the same direction and speed as the user of the mobile phone terminal device, the mobile phone terminal device is in the detection area. Has a problem of detecting this as an obstacle.

  This invention is made in view of this point, and it aims at providing the portable terminal device and the collision danger determination alerting | reporting method which can detect the obstacle which may collide quickly with sufficient precision. .

  The mobile terminal device according to claim 1, a camera that captures a front whole image and generates whole image data, and extracts object contour data of the whole image data from the camera to extract object contour data. A contour extracting means for generating, a feature detecting means for detecting a feature amount of the object contour data of the contour extracting means, an obstacle is recognized based on the feature amount of the object contour data from the feature detecting means, and Collision risk determination means for predicting the motion of the object contour data and determining the risk of collision to generate a collision risk level, and reporting the risk of collision when the collision risk level from the collision risk determination means is greater than or equal to a threshold value And a collision risk notification means.

  According to this configuration, the feature amount of the object contour data is detected, the obstacle is recognized based on the feature amount of the object contour data, the motion of the object contour data is predicted, the collision risk is determined, and the collision risk is determined. Since the degree is generated, an obstacle that may collide can be detected early with high accuracy.

  According to a second aspect of the present invention, there is provided the portable terminal device according to the first aspect, wherein the collision risk determination means includes an obstacle feature database having obstacle feature data which is feature quantity data of the obstacle, and the feature. Obstacle recognition means for comparing the feature quantity detected by the detection means with the obstacle feature data in the obstacle feature database to recognize obstacles and generate obstacle recognition data; and the feature detection means A motion detection / prediction unit that detects a motion of the object image data based on a temporal change of the feature amount detected by the step and predicts a motion to generate motion prediction data; and the obstacle recognition data and the motion prediction data. A risk determination unit that determines the risk of collision based on the generated risk and generates a risk of collision.

  According to this configuration, in addition to the effect of the first aspect of the invention, the accuracy of detecting an obstacle that may collide can be further improved.

  The collision risk determination / notification method according to claim 3, wherein a whole image data generating step of photographing a whole front image to generate whole image data, and extracting an outline of the object image data of the whole image data to extract the object outline A contour extracting step for generating data, a feature detecting step for detecting a feature amount of the object contour data, an obstacle is recognized based on the feature amount of the object contour data, and a motion of the object contour data is predicted. A collision risk determination step for determining a collision risk and generating a collision risk level, and a danger notification step for notifying the collision risk when the collision risk level in the risk determination step is equal to or greater than a threshold value. did.

  According to this method, the feature amount of the object contour data is detected, the obstacle is recognized based on the feature amount of the object contour data, the motion of the object contour data is predicted, the collision risk is determined, and the collision risk is determined. Since the degree is generated, an obstacle that may collide can be detected early with high accuracy.

  According to a fourth aspect of the present invention, in the collision risk determination notification method according to the third aspect of the present invention, the collision risk determination step includes the feature amount detected in the feature detection step and obstacle feature data in the obstacle feature database. An obstacle recognition step for generating obstacle recognition data by comparing obstacles with each other, and detecting the movement of the object image data based on the temporal change of the feature amount detected in the feature detection step A motion detection prediction step for predicting the motion and generating motion prediction data; and a risk determination step for determining a risk of collision based on the obstacle recognition data and the motion prediction data to generate a collision risk level. It was made to have.

  According to this method, in addition to the effect of the invention described in claim 3, it is possible to further improve the accuracy of detecting an obstacle that may collide.

  As described above, according to the present invention, the feature amount of the object contour data is detected, the obstacle is recognized based on the feature amount of the object contour data, and the motion of the object contour data is predicted to predict the risk of collision. Therefore, it is possible to detect obstacles that may collide with high accuracy at an early stage.

  The gist of the present invention is to detect a feature amount of object contour data, recognize an obstacle based on the feature amount of the object contour data, and predict a motion of the object contour data to determine a collision risk, thereby determining a collision risk. Is to generate degrees.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(One embodiment)
FIG. 1 is a block diagram showing a configuration of a mobile terminal device according to an embodiment of the present invention. As shown in FIG. 1, a mobile terminal device 100 according to an embodiment of the present invention includes an antenna 101, a communication unit 102, a camera 103, a contour extraction unit 104, a feature detection unit 105, an obstacle feature database 106, and an obstacle. A recognition unit 107, a motion detection prediction unit 108, a danger determination unit 109, an alarm signal generation unit 110, a signal data processing unit 111, a danger notification unit 112, a display unit 113, an operation unit 114, a reception unit 115, and a transmission unit 116 are provided. doing.

  The antenna 101 transmits transmission data from the communication unit 102 as a radio signal. Further, the antenna 101 receives data from the communication apparatus of the communication partner, generates reception data, and gives it to the communication unit 102. The communication unit 102 sends reception data from the antenna 101 to the signal data processing unit 111. The signal data processing unit 111 performs predetermined processing on the reception data from the antenna 101 and sends the reception data to the reception unit 115 or the display unit 113. The signal data processing unit 111 performs predetermined processing on the transmission data from the transmission unit 116 and transmits the transmission data to the antenna 101 via the communication unit 102. The operation unit 114 is for a user to input an operation signal to the signal data processing unit 111.

  The camera 103 shoots the entire front image, generates the entire image data, and gives it to the contour extraction unit 104. When the collision warning function is activated, the camera 103 continuously shoots even in a state where an e-mail operation, a browsing operation or a game operation when connected to the Internet is performed, and the user of the mobile terminal device The whole front image is photographed to generate whole image data.

  The contour extraction unit 104 extracts the contour of the object image data of the entire image data from the camera 103 to generate the object contour data, and gives it to the feature detection unit 105. The feature detection unit 105 detects the feature amount of the object contour data from the contour extraction unit 104 and supplies it to the obstacle recognition unit 107 and the motion detection prediction unit 108.

  The obstacle feature database 106 has obstacle feature data that is data of feature quantities of obstacles such as people, bicycles, and utility poles in advance. The obstacle recognition unit 107 compares the feature amount detected by the feature detection unit 105 with the obstacle feature data in the obstacle feature database 106, recognizes the obstacle, generates obstacle recognition data, and determines the risk. Part 109. The motion detection / prediction unit 108 detects the motion of the object image data based on the temporal change of the feature amount detected by the feature detection unit 105, predicts the motion, generates motion prediction data, and gives the motion determination unit 109. The danger determination unit 109 determines a collision risk based on the obstacle recognition data from the obstacle recognition unit 107 and the motion prediction data from the motion detection prediction unit 108, generates a collision risk, and generates an alarm signal generation unit 110. And the signal data processing unit 111. The alarm signal generation unit 110 generates an alarm signal and gives it to the signal data processing unit 111 when the risk of collision from the danger determination unit 109 is greater than or equal to a threshold value. The signal data processing unit 111 gives a drive signal to the danger notification unit 112 when receiving the warning signal from the warning signal generation unit 110. The danger notifying unit 112 notifies the danger of a collision when receiving a driving signal from the signal data processing unit 111. The danger notification unit 112 is constituted by, for example, an LED display, and notifies the danger of collision by blinking the LED display.

  Further, the signal data processing unit 111 displays a danger of collision on the display unit 113 when receiving an alarm signal from the alarm signal generation unit 110, and causes the receiving unit 115 to notify the danger of collision with sound.

  Next, the operation of the mobile terminal device 100 according to the embodiment of the present invention will be described with reference to FIG. 2 together with FIG.

  In step ST101, the camera 103 captures the entire front image and generates entire image data. Next, in step ST102, the contour extracting unit 104 extracts the contour of the object image data of the entire image data from the camera 103, and generates object contour data.

  Next, in step ST <b> 103, the feature detection unit 105 detects the feature amount of the object contour data from the contour extraction unit 104. Next, in step ST104, the obstacle recognition unit 107 compares the feature amount detected by the feature detection unit 105 with the obstacle feature data in the obstacle feature database 106, recognizes the obstacle, and recognizes the obstacle. Generate data. At the same time, in step ST105, the motion detection / prediction unit 108 detects the motion of the object image data based on the temporal change of the feature amount detected by the feature detection unit 105, predicts the motion, and generates motion prediction data. .

  Next, in step ST106, the danger determination unit 109 determines the risk of collision based on the obstacle recognition data from the obstacle recognition unit 107 and the motion prediction data from the motion detection prediction unit 108, and generates a collision risk degree. To do. Next, in step ST107, the alarm signal generation unit 110 determines whether or not the collision risk from the danger determination unit 109 is equal to or greater than a threshold value. When the collision risk is greater than or equal to the threshold value in step ST107, the alarm signal generator 110 generates an alarm signal in step ST108.

  Next, in step ST109, when the signal data processing unit 111 receives an alarm signal from the alarm signal generation unit 110, the signal data processing unit 111 gives a driving signal to the danger notification unit 112 to notify the danger of collision. When the collision risk from the danger determination unit 109 is not equal to or greater than the threshold value in step ST107, the process goes to step ST110. Next, in step ST110, the signal data processing unit 111 determines whether or not it is an end selection. In step ST110, the signal data processing unit 111 returns to step ST101 when the end selection is not made.

  In one embodiment of the present invention, the danger notification unit 112 receives information on any one of the distance, direction, speed, and type of an obstacle at risk of collision, or information that is a combination of these pieces of information. The user of the mobile terminal device may be notified using either or both of the image and the sound. Moreover, in one embodiment of the present invention, the danger notification unit 112 may be configured to notify by vibration using a vibration device (vibrator). The present invention can also be applied to a portable information terminal device such as a PDA.

  The present invention is useful for determining and reporting a collision risk.

The block diagram which shows the structure of the portable terminal device which concerns on one embodiment of this invention The flowchart for demonstrating operation | movement of the portable terminal device which concerns on one embodiment of this invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Portable terminal device 101 Antenna 102 Communication part 103 Camera 104 Contour extraction part 105 Feature detection part 106 Obstacle feature database 107 Obstacle recognition part 108 Motion detection prediction part 109 Risk judgment part 110 Warning signal generation part 111 Signal data processing part 112 Risk Notification unit 113 Display unit 114 Operation unit 115 Reception unit 116 Transmission unit

Claims (4)

  A camera that captures a front whole image and generates whole image data, a contour extraction unit that extracts a contour of object image data of the whole image data from the camera and generates object contour data, and the contour extraction unit Detecting a feature amount of the object contour data; recognizing an obstacle based on the feature amount of the object contour data from the feature detection unit; and predicting a motion of the object contour data to detect a collision A mobile device comprising: a collision risk determination unit that determines a risk and generates a collision risk level; and a collision risk notification unit that reports a collision risk when the collision risk level from the collision risk determination unit is greater than or equal to a threshold value. Terminal device.   The collision risk determination means includes an obstacle feature database having obstacle feature data that is data of an obstacle feature quantity, the feature quantity detected by the feature detection means, and the obstacle feature of the obstacle feature database. Obstacle recognition means for comparing the data with each other to generate obstacle recognition data, and the movement of the object image data based on the temporal change of the feature amount detected by the feature detection means. A motion detection / prediction unit that detects motion and predicts motion to generate motion prediction data; a risk determination unit that determines the risk of collision based on the obstacle recognition data and the motion prediction data; The mobile terminal device according to claim 1, comprising:   A whole image data generating step of capturing a whole front image and generating whole image data; a contour extracting step of generating object contour data by extracting a contour of the object image data of the whole image data; and the object contour data A feature detection step of detecting a feature amount of the object, and an obstacle is recognized based on the feature amount of the object contour data, and a collision risk is determined by predicting a motion of the object contour data to determine a collision risk. A collision risk determination notification method comprising: a collision risk determination step; and a risk notification step of notifying a collision risk when the collision risk level in the risk determination step is equal to or greater than a threshold value.   The collision risk determination step compares the feature amount detected in the feature detection step with the obstacle feature data in the obstacle feature database, recognizes the obstacle, and generates obstacle recognition data. A motion detection prediction step for detecting motion of the object image data based on a temporal change of the feature amount detected in the feature detection step and predicting motion to generate motion prediction data, and the obstacle recognition The collision risk determination notification method according to claim 3, further comprising: a risk determination step of determining a collision risk based on the data and the motion prediction data to generate a collision risk level.

Images