CN105319046A - Method of collision detection and alarm and portable device of collision detection and alarm - Google Patents
Method of collision detection and alarm and portable device of collision detection and alarm Download PDFInfo
- Publication number
- CN105319046A CN105319046A CN201410375056.5A CN201410375056A CN105319046A CN 105319046 A CN105319046 A CN 105319046A CN 201410375056 A CN201410375056 A CN 201410375056A CN 105319046 A CN105319046 A CN 105319046A
- Authority
- CN
- China
- Prior art keywords
- driver
- edge
- shooting
- image
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention provides a method of collision detection and alarm and a portable device of collision detection and alarm. The method of collision detection and alarm comprises: obtaining acceleration magnitudes of a vehicle in x, y and z directions; obtaining a total acceleration magnitude according to the acceleration magnitudes of the vehicle in x, y and z directions; determining whether the total acceleration magnitude is larger than a preset acceleration magnitude or not; shooting driver's images when the total acceleration magnitude is larger than the preset acceleration magnitude; comparing the driver's images with a preset safe driving feature database, and determining whether the driver is in a safe state or not; and sending alarm information when the driver is not in the safe state. According to the invention, the method of collision detection and alarm may be realized through a portable device containing corresponding hardware; and moreover, the method may determine collision and call for help immediately when an accident happens to a vehicle in the prior art not provided with a special alarm device.
Description
Technical field
The present invention relates to communication technical field, be specifically related to method and the mancarried device of a kind of collision detection and warning.
Background technology
Along with automobile extensively popularizing in society, motor-vehicle accident frequently occurs, and China is annual because the number of motor-vehicle accident death is up to more than 100,000.A lot of accident, because the wounded lies in a comatose condition, can not notify rescue centre timely and cause death or the condition of the injury to increase the weight of, and how judges collision in the very first time that has an accident, and notifies rescue centre, for significant the wounded.
Current most accident alarming method is all carry out collision detection based on the software and hardware that automobile is built-in specially, although its accuracy is higher, expensive, and because its erected cost is higher, many automobiles are not all installed.
How when not installing special warning device in automobile, in having an accident, the very first time judges collision, and sends help information, becomes the problem needing now to wait to solve.
Summary of the invention
In order to solve when not installing special warning device in automobile, in having an accident, the very first time judges collision, and sends help information, and the invention provides a kind of method of collision detection and warning, described method comprises:
Obtain the accekeration of vehicle in x, y, z all directions;
According to the accekeration of described vehicle in x, y, z all directions obtained, obtain a total acceleration value;
Judge whether described total acceleration value is greater than a predetermined acceleration value;
When described total acceleration value is greater than described predetermined acceleration value, shooting driver image;
Described driver's image and a preset security driving characteristics storehouse are contrasted, judges that whether described driver is in a safe condition;
When described driver is not in described safe condition, send warning message.
Preferably, described method also comprises:
When described total acceleration value is less than or equal to described predetermined acceleration value, obtain the angular acceleration values of described vehicle in x, y, z all directions;
According to the angular acceleration values of the described vehicle of acquisition in x, y, z all directions, obtain a total angular acceleration values;
Judge whether described total angular acceleration values is greater than a preset angle accekeration;
When described total angular acceleration values is greater than described preset angle accekeration, perform shooting driver image step;
If not, return execution and obtain the accekeration step of described vehicle in x, y, z all directions.
Preferably, the step of described shooting driver image also comprises:
Obtain the distance value with object in shooting direction;
Judge whether described distance value is less than or equal to a predeterminable range value;
When described distance value is greater than a predeterminable range value, perform described shooting driver image step;
When described distance value is less than or equal to a predeterminable range value, perform described transmission warning message step.
Preferably, described method also comprises the step setting up described preset security driving characteristics storehouse, and described step comprises:
Driver's image under acquisition safe condition;
Extract the edge of driver's driver's image under described safe condition, set up one first edge histogram;
Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram;
Described first edge orientation histogram is saved to described preset security driving characteristics storehouse.
Preferably, describedly by the step that described driver's image and a preset security driving characteristics storehouse carry out contrasting be:
Extract driver at the edge of described driver's image, set up one second edge histogram,
Extract the gradient direction of each marginal point on described second edge histogram, set up one second edge orientation histogram;
Described first edge orientation histogram in described second edge orientation histogram and described preset security driving characteristics storehouse is contrasted, when containing described first edge orientation histogram identical with described second edge orientation histogram in described preset security driving characteristics storehouse, then judge that described driver is in described safe condition; If not, then judge that described driver is not in described safe condition.
Present invention also offers the mancarried device of a kind of collision detection and warning, described mancarried device comprises:
One gravity sensing unit, for obtaining the accekeration of vehicle in x, y, z all directions;
One processing unit, for according to the accekeration of described vehicle in x, y, z all directions obtained, obtains a total acceleration value; Judge whether described total acceleration value is greater than a predetermined acceleration value;
One shooting unit, during for judging that described total acceleration value is greater than described predetermined acceleration value when described processing unit, shooting driver image;
One storage unit, for storing a preset security driving characteristics storehouse; Described processing unit also for described driver's image and a preset security driving characteristics storehouse being contrasted, judges that whether described driver is in a safe condition;
One alarm unit, for when described processing unit judges that described driver is not in described safe condition, sends warning message.
Preferably, described mancarried device also comprises:
One angular velocity sensing unit, during for judging that described total acceleration value is less than or equal to described predetermined acceleration value when described processing unit, obtains the angular acceleration values of described vehicle in x, y, z all directions;
Described processing unit also for according to the angular acceleration values of the described vehicle of acquisition in x, y, z all directions, obtains a total angular acceleration values; Judge whether described total angular acceleration values is greater than a preset angle accekeration; When described total angular acceleration values is greater than described preset angle accekeration, sends shooting instruction to described shooting unit and perform shooting driver image step; When described total angular acceleration values is less than or equal to described preset angle accekeration, sends instruction to described gravity sensing unit and perform the accekeration step of the described vehicle of acquisition in x, y, z all directions.
Preferably, described mancarried device also comprises a distance sensing unit, for obtaining the distance value with object in shooting direction;
Described processing unit is for judging whether described distance value is less than or equal to a predeterminable range value; When described distance value is greater than a predeterminable range value, sends shooting instruction and perform described shooting driver image step to described shooting unit; When described distance value is less than or equal to a predeterminable range value, sends instruction to described alarm unit and perform described transmission warning message step.
Preferably, described mancarried device also comprises a graphics processing unit, for according to driver's image under the safe condition of described shooting unit acquisition, extracts the edge of driver's driver's image under described safe condition, sets up one first edge histogram; Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram; Described first edge orientation histogram is saved to the described preset security driving characteristics storehouse in described storage unit.
Preferably, the described driver image of described graphics processing unit also for obtaining according to described shooting unit, extract driver at the edge of described driver's image, set up one second edge histogram, extract the gradient direction of each marginal point on described second edge histogram, set up one second edge orientation histogram;
Described processing unit is also for contrasting described first edge orientation histogram in described second edge orientation histogram and described preset security driving characteristics storehouse, when containing described first edge orientation histogram identical with described second edge orientation histogram in described preset security driving characteristics storehouse, then judge that described driver is in described safe condition; If not, then judge that described driver is not in described safe condition.
As mentioned above, the method for a kind of collision detection provided by the invention and warning and mancarried device, have following beneficial effect:
1) by gravity sensor and shooting driver image, the accuracy judging collision is added.
2) when vehicle collides in front, the numerical value of its total acceleration value can be comparatively large, but when the side of vehicle is subject to clashing into, the numerical value of its total acceleration value changes can't be too large.The present invention by obtaining total magnitude of angular velocity, and judges whether to be subject to leaning to one side to clash into according to described total magnitude of angular velocity, avoids cannot being detected problem when side is knocked.
3) if mancarried device is knocked to falling ground, camera possibly cannot obtain driver's image.Whether the present invention answers unit judges device to be shifted by distance perspective, to avoid when driver's image cannot be obtained after device is knocked cannot and the problem of alarm.
4) consider the problem of data processing speed and power consumption, use the method for Shape-based interpolation edge orientation histogram to realize the differentiation of safe condition, accelerate the speed of data processing, decrease the power consumption of device.
5) realized the method for crash tests pre-alarm provided by the invention by mancarried device, solve prior art as do not installed special warning device at automotive interior, then cannot realize the problem of crash tests and warning.And of the prior artly portablely generally all contain the hardware that acceleration of gravity inductor, distance-sensor, angular velocity inductor etc. realize the inventive method, without the need to adding hardware device again, therefore without the need to increasing extra cost.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of method one embodiment of a kind of collision detection of the present invention and warning;
Fig. 2 is the idiographic flow schematic diagram of step S4 in method one embodiment of a kind of collision detection of the present invention and warning;
Fig. 3 is the idiographic flow schematic diagram of step S5 in method one embodiment of a kind of collision detection of the present invention and warning;
Fig. 4 is the structural representation of mancarried device one embodiment of a kind of collision detection of the present invention and warning.
Embodiment
Technical scheme of the present invention is understood better in order to make relevant technical staff in the field, below in conjunction with the accompanying drawing of embodiment of the present invention, technical scheme in embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.
The invention provides a kind of method of collision detection and warning, as shown in Figure 1, described method comprises:
Step S1: obtain the accekeration of vehicle in x, y, z all directions.
Obtain the accekeration of vehicle in the process of moving in all directions, generally obtain by gravity sensor.
Step S2: according to the accekeration of described vehicle in x, y, z all directions obtained, obtain a total acceleration value.
The computing formula of total acceleration value is:
wherein Tx1, Ty1, Tz1 represent gravity sensor x respectively, the acceleration in y, z direction, and Ttotal1 is total acceleration value.
Step S3: judge whether total acceleration value is greater than a predetermined acceleration value.
In general, all close at the uniform velocity travelling (acceleration is less than 1g) as car crass is last, can produce multiple acceleration impact signal in the moment of head-on crash, maximum impact signal is generally all more than 10g.Preferably, predetermined acceleration value can be set to 8g by user.
When described total acceleration value is less than or equal to described predetermined acceleration value, perform step S31: obtain the angular acceleration values of vehicle in x, y, z all directions.
Step S32: according to the angular acceleration values of acquisition vehicle in x, y, z all directions, obtain a total angular acceleration values.
The computing formula of total angular acceleration values is:
wherein Tx2, Ty2, Tz2 represent gyroscope x respectively, the angular acceleration in y, z direction, and Ttotal2 is total angular acceleration values.
Step S32: judge whether total angular acceleration values is greater than a preset angle accekeration.
When vehicle is at side collision, accekeration is compared with little during head-on crash, but automobile can produce larger angular deflection.The present invention, by increasing the scheme of the magnitude of angular velocity obtained in vehicle traveling in each orientation, judges that whether total angular acceleration values is more than a preset angle accekeration, avoids omitting the situation differentiating that side is hit, and incurs loss through delay the situation of the time of reporting to the police.Preferably, preset angle accekeration is per second square of π/3.
It should be noted that, in other embodiments, judge that the angular deflection of automobile also can judge by direction sensor the angle that automobile offsets.
When described total angular acceleration values is less than or equal to described preset angle accekeration, return step S1: perform and obtain the accekeration of vehicle in x, y, z all directions.
When described total angular acceleration values is greater than described preset angle accekeration, perform step S4: perform shooting driver image.
When described total acceleration value is greater than described predetermined acceleration value, same also execution step S4: shooting driver image.
It should be noted that, method provided by the invention is the orientation that driver, device fixed placement can be photographed driver, and can not the position of movement in normal driving conditions.
Driver's image of shooting can be one, or multiple.
When general, device is fixed on the position that can photograph driver, if but device is knocked to when falling ground, and camera possibly cannot obtain driver's image.
Whether the present invention also answers unit judges device to be shifted by providing by distance perspective, to avoid when driver's image cannot be obtained after device is knocked cannot and the problem of alarm.
Fig. 2 is the idiographic flow schematic diagram of step S4 in method one embodiment of a kind of collision detection of the present invention and warning, and as shown in Figure 2, described step S4 comprises the steps:
Step S41: obtain the distance value with object in shooting direction;
Step S42: whether judging distance value is less than or equal to a predeterminable range value;
When described distance value is greater than a predeterminable range value, perform step S43: shooting driver image;
When described distance value is less than or equal to a predeterminable range value, perform step S6: send warning message.
When after acquisition driver image, perform step S5: driver's image and a preset security driving characteristics storehouse are contrasted, judge that whether driver is in a safe condition.
When driver normally drives, upper body keeps substantially vertical state with seat, and after getting into an accident, health generally larger angle of inclination can occur, and the global shape of people there occurs larger change.Compared by driver's image and the driver's image in the safe condition in the safe driving feature database pre-set, then can judge that whether driver is in a safe condition faster.
Consider the problem of data processing speed and power consumption, preferably, present invention also offers the step setting up described preset security driving characteristics storehouse, described step comprises (not shown in FIG.):
Driver's image under acquisition safe condition;
Extract the edge of driver's driver's image under described safe condition, set up one first edge histogram;
Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram;
Described first edge orientation histogram is saved to described preset security driving characteristics storehouse.
On the above-mentioned basis setting up the step in described preset security driving characteristics storehouse, accordingly, as shown in Figure 3, describedly by the step that described driver's image and a preset security driving characteristics storehouse carry out contrasting be:
Step S51: extract driver at the edge of driver's image, set up one second edge histogram.
Step S52: the gradient direction extracting each marginal point on the second edge histogram, sets up one second edge orientation histogram.
Step S53: described first edge orientation histogram in the second edge orientation histogram and preset security driving characteristics storehouse is contrasted, when containing first edge orientation histogram identical with the second edge orientation histogram in preset security driving characteristics storehouse, then judge that driver is in described safe condition; If not, then judge that driver is not in described safe condition.
When described driver is not in described safe condition, perform step S6: send warning message.
Sending warning message can for be sent to the information content comprising current location the programmed emergency pre-set; Also can directly dial programmed emergency phone, play the voice help-asking information of editing in advance; Also can sound the alarm, obtain the relief of people near the scene of the accident.
Present invention also offers the mancarried device of a kind of collision detection and warning, described mancarried device comprises as shown in Figure 4: gravity sensing unit 1, processing unit 2, shooting unit 3, storage unit 4, alarm unit 5, angular velocity sensing unit 6, distance sensing unit 7, graphics processing unit 8.
Wherein, gravity sensing unit 1, for obtaining the accekeration of vehicle in x, y, z all directions.
Processing unit 2, is electrically connected gravity sensing unit 1, for according to the accekeration of described vehicle in x, y, z all directions obtained, obtains a total acceleration value; Judge whether described total acceleration value is greater than a predetermined acceleration value.
One shooting unit 3, is electrically connected process unit 2, for when described processing unit 3 judges that described total acceleration value is greater than described predetermined acceleration value, and shooting driver image.
One storage unit 4, for storing a preset security driving characteristics storehouse; Described processing unit 2 is also electrically connected storage unit 4, for described driver's image and a preset security driving characteristics storehouse being contrasted, judges that whether described driver is in a safe condition.
One alarm unit 5, is electrically connected process unit 2, for when described process single 2 judges that described driver is not in described safe condition, sends warning message.
Preferably, described mancarried device also comprises: an angular velocity sensing unit 6, be electrically connected process unit 2, for when described processing unit 2 judges that described total acceleration value is less than or equal to described predetermined acceleration value, obtain the angular acceleration values of described vehicle in x, y, z all directions.
Described process single 2 also for according to the angular acceleration values of the described vehicle of acquisition in x, y, z all directions, obtains a total angular acceleration values; Judge whether described total angular acceleration values is greater than a preset angle accekeration; When described total angular acceleration values is greater than described preset angle accekeration, sends shooting instruction to described shooting unit 3 and perform shooting driver image step; When described total angular acceleration values is less than or equal to described preset angle accekeration, sends instruction to described gravity sensing unit 1 and perform the accekeration step of the described vehicle of acquisition in x, y, z all directions.
Preferably, described mancarried device also comprises a distance sensing unit 7, and electric wire connection point reason unit 2, for obtaining the distance value with object in shooting direction.
Described processing unit 2 is for judging whether described distance value is less than or equal to a predeterminable range value; When described distance value is greater than a predeterminable range value, sends shooting instruction and perform described shooting driver image step to described shooting unit 3; When described distance value is less than or equal to a predeterminable range value, sends instruction to described alarm unit 5 and perform described transmission warning message step.
Preferably, described mancarried device also comprises a graphics processing unit 8, be electrically connected shooting unit 3 and storage unit 4, for driver's image under the safe condition that obtains according to described shooting unit 3, extract the edge of driver's driver's image under described safe condition, set up one first edge histogram; Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram; Described first edge orientation histogram is saved to the described preset security driving characteristics storehouse in described storage unit 4.
Preferably, the described driver image of described graphics processing unit 8 also for obtaining according to described shooting unit 3, extract driver at the edge of described driver's image, set up one second edge histogram, extract the gradient direction of each marginal point on described second edge histogram, set up one second edge orientation histogram;
Described processing unit 2 is also electrically connected graphics processing unit 8, for described first edge orientation histogram in described second edge orientation histogram and described preset security driving characteristics storehouse is contrasted, when containing described first edge orientation histogram identical with described second edge orientation histogram in described preset security driving characteristics storehouse, then judge that described driver is in described safe condition; If not, then judge that described driver is not in described safe condition.
The mancarried device that it should be noted that a kind of collision detection provided by the invention and warning can be smart mobile phone of the prior art, panel computer, portable GPS device etc., can also be other mancarried device certainly.
In sum, the method for a kind of collision detection provided by the invention and warning and mancarried device, have following beneficial effect:
1) by gravity sensor and shooting driver image, the accuracy judging collision is added.
2) when vehicle collides in front, the numerical value of its total acceleration value can be comparatively large, but when the side of vehicle is subject to clashing into, the numerical value of its total acceleration value changes can't be too large.The present invention by obtaining total magnitude of angular velocity, and judges whether to be subject to leaning to one side to clash into according to described total magnitude of angular velocity, avoids cannot being detected problem when side is knocked.
3) if device is knocked to falling ground, camera possibly cannot obtain driver's image.Whether the present invention answers unit judges device to be shifted by distance perspective, to avoid when driver's image cannot be obtained after device is knocked cannot and the problem of alarm.
4) consider the problem of data processing speed and power consumption, use the method for Shape-based interpolation edge orientation histogram to realize the differentiation of safe condition, accelerate the speed of data processing, decrease the power consumption of device.
5) realized the method for crash tests pre-alarm provided by the invention by mancarried device, solve prior art as do not installed special warning device at automotive interior, then cannot realize the problem of crash tests and warning.And of the prior artly portablely generally all contain the hardware that acceleration of gravity inductor, distance-sensor, angular velocity inductor etc. realize the inventive method, without the need to adding hardware device again, therefore without the need to increasing extra cost.
Below only have expressed several embodiment of the present invention, it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. a method for collision detection and warning, is characterized in that, described method comprises:
Obtain the accekeration of vehicle in x, y, z all directions;
According to the accekeration of described vehicle in x, y, z all directions obtained, obtain a total acceleration value;
Judge whether described total acceleration value is greater than a predetermined acceleration value;
When described total acceleration value is greater than described predetermined acceleration value, shooting driver image;
Described driver's image and a preset security driving characteristics storehouse are contrasted, judges that whether described driver is in a safe condition;
When described driver is not in described safe condition, send warning message.
2. the method for collision detection and warning according to claim 1, is characterized in that, described method also comprises:
When described total acceleration value is less than or equal to described predetermined acceleration value, obtain the angular acceleration values of described vehicle in x, y, z all directions;
According to the angular acceleration values of the described vehicle of acquisition in x, y, z all directions, obtain a total angular acceleration values;
Judge whether described total angular acceleration values is greater than a preset angle accekeration;
When described total angular acceleration values is greater than described preset angle accekeration, perform shooting driver image step;
If not, return execution and obtain the accekeration step of described vehicle in x, y, z all directions.
3. the method for collision detection according to claim 1 and warning, is characterized in that, the step of described shooting driver image also comprises:
Obtain the distance value with object in shooting direction;
Judge whether described distance value is less than or equal to a predeterminable range value;
When described distance value is greater than a predeterminable range value, perform described shooting driver image step;
When described distance value is less than or equal to a predeterminable range value, perform described transmission warning message step.
4. the method for collision detection according to claim 1 and warning, is characterized in that, described method also comprises the step setting up described preset security driving characteristics storehouse, and described step comprises:
Driver's image under acquisition safe condition;
Extract the edge of driver's driver's image under described safe condition, set up one first edge histogram;
Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram;
Described first edge orientation histogram is saved to described preset security driving characteristics storehouse.
5. the method for collision detection according to claim 4 and warning, is characterized in that, describedly by the step that described driver's image and a preset security driving characteristics storehouse carry out contrasting is:
Extract driver at the edge of described driver's image, set up one second edge histogram;
Extract the gradient direction of each marginal point on described second edge histogram, set up one second edge orientation histogram;
Described first edge orientation histogram in described second edge orientation histogram and described preset security driving characteristics storehouse is contrasted, when containing described first edge orientation histogram identical with described second edge orientation histogram in described preset security driving characteristics storehouse, described driver is for being in described safe condition; If not, described driver is not for be in described safe condition.
6. a mancarried device for collision detection and warning, is characterized in that, described mancarried device comprises:
One gravity sensing unit, for obtaining the accekeration of vehicle in x, y, z all directions;
One processing unit, for according to the accekeration of described vehicle in x, y, z all directions obtained, obtains a total acceleration value; Judge whether described total acceleration value is greater than a predetermined acceleration value;
One shooting unit, during for judging that described total acceleration value is greater than described predetermined acceleration value when described processing unit, shooting driver image;
One storage unit, for storing a preset security driving characteristics storehouse; Described processing unit also for described driver's image and a preset security driving characteristics storehouse being contrasted, judges that whether described driver is in a safe condition;
One alarm unit, for when described processing unit judges that described driver is not in described safe condition, sends warning message.
7. the mancarried device of collision detection and warning according to claim 6, is characterized in that, described mancarried device also comprises:
One angular velocity sensing unit, during for judging that described total acceleration value is less than or equal to described predetermined acceleration value when described processing unit, obtains the angular acceleration values of described vehicle in x, y, z all directions;
Described processing unit also for according to the angular acceleration values of the described vehicle of acquisition in x, y, z all directions, obtains a total angular acceleration values; Judge whether described total angular acceleration values is greater than a preset angle accekeration; When described total angular acceleration values is greater than described preset angle accekeration, sends shooting instruction to described shooting unit and perform shooting driver image step; When described total angular acceleration values is less than or equal to described preset angle accekeration, sends instruction to described gravity sensing unit and perform the accekeration step of the described vehicle of acquisition in x, y, z all directions.
8. the mancarried device of collision detection according to claim 6 and warning, is characterized in that, described mancarried device also comprises a distance sensing unit, for obtaining the distance value with object in shooting direction;
Described processing unit is for judging whether described distance value is less than or equal to a predeterminable range value; When described distance value is greater than a predeterminable range value, sends shooting instruction and perform described shooting driver image step to described shooting unit; When described distance value is less than or equal to a predeterminable range value, sends instruction to described alarm unit and perform described transmission warning message step.
9. the mancarried device of collision detection according to claim 6 and warning, it is characterized in that, described mancarried device also comprises a graphics processing unit, for driver's image under the safe condition that obtains according to described shooting unit, extract the edge of driver's driver's image under described safe condition, set up one first edge histogram; Extract the gradient direction of each marginal point on described first edge histogram, set up one first edge orientation histogram; Described first edge orientation histogram is saved to the described preset security driving characteristics storehouse in described storage unit.
10. the mancarried device of collision detection according to claim 9 and warning, it is characterized in that, the described driver image of described graphics processing unit also for obtaining according to described shooting unit, extract driver at the edge of described driver's image, set up one second edge histogram, extract the gradient direction of each marginal point on described second edge histogram, set up one second edge orientation histogram;
Described processing unit is also for contrasting described first edge orientation histogram in described second edge orientation histogram and described preset security driving characteristics storehouse, when containing described first edge orientation histogram identical with described second edge orientation histogram in described preset security driving characteristics storehouse, then judge that described driver is in described safe condition; If not, then judge that described driver is not in described safe condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410375056.5A CN105319046A (en) | 2014-07-30 | 2014-07-30 | Method of collision detection and alarm and portable device of collision detection and alarm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410375056.5A CN105319046A (en) | 2014-07-30 | 2014-07-30 | Method of collision detection and alarm and portable device of collision detection and alarm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105319046A true CN105319046A (en) | 2016-02-10 |
Family
ID=55246919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410375056.5A Pending CN105319046A (en) | 2014-07-30 | 2014-07-30 | Method of collision detection and alarm and portable device of collision detection and alarm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105319046A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108932836A (en) * | 2018-08-02 | 2018-12-04 | 安徽江淮汽车集团股份有限公司 | A kind of distant collision alarm method and system |
| CN109649316A (en) * | 2018-11-28 | 2019-04-19 | 江苏大学 | A kind of distress system and method based on vehicle collision class accident |
| CN110223503A (en) * | 2019-06-10 | 2019-09-10 | 兰州交通大学 | A kind of expressway traffic accident early warning and autonomous method for seeking help |
| CN110276932A (en) * | 2019-06-24 | 2019-09-24 | 上海擎感智能科技有限公司 | A kind of emergency backup methods, device and computer-readable medium for vehicle |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1420036A (en) * | 2001-11-16 | 2003-05-28 | 中国科学院力学研究所 | Ramp automobile collision experiment system and method |
| KR20080012612A (en) * | 2006-08-04 | 2008-02-12 | 엘지전자 주식회사 | Apparatus and method for transmitting sos message in car navigation device |
| CN101857008A (en) * | 2010-04-29 | 2010-10-13 | 路军 | Active-mode intelligent safety monitoring instrument for car |
-
2014
- 2014-07-30 CN CN201410375056.5A patent/CN105319046A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1420036A (en) * | 2001-11-16 | 2003-05-28 | 中国科学院力学研究所 | Ramp automobile collision experiment system and method |
| KR20080012612A (en) * | 2006-08-04 | 2008-02-12 | 엘지전자 주식회사 | Apparatus and method for transmitting sos message in car navigation device |
| CN101857008A (en) * | 2010-04-29 | 2010-10-13 | 路军 | Active-mode intelligent safety monitoring instrument for car |
Non-Patent Citations (2)
| Title |
|---|
| 徐叶雷等: ""基于MEMS加速度计的车用自动呼救系统"", 《传感技术学报》 * |
| 郑培: ""机动车驾驶员驾驶疲劳测评方法的研究"", 《中国优秀博硕士学位论文全文数据库（博士） 工程科技II辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108932836A (en) * | 2018-08-02 | 2018-12-04 | 安徽江淮汽车集团股份有限公司 | A kind of distant collision alarm method and system |
| CN109649316A (en) * | 2018-11-28 | 2019-04-19 | 江苏大学 | A kind of distress system and method based on vehicle collision class accident |
| CN110223503A (en) * | 2019-06-10 | 2019-09-10 | 兰州交通大学 | A kind of expressway traffic accident early warning and autonomous method for seeking help |
| CN110276932A (en) * | 2019-06-24 | 2019-09-24 | 上海擎感智能科技有限公司 | A kind of emergency backup methods, device and computer-readable medium for vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109278757B (en) | Vehicle collision detection method and vehicle control system | |
| CN105096543B (en) | A kind of vehicular rescue method and system based on drive recorder | |
| Bo et al. | You're driving and texting: detecting drivers using personal smart phones by leveraging inertial sensors | |
| EP2481037B1 (en) | Driver assistance device and system for vehicle accident detection and method for detecting a vehicle accident | |
| CN103568945B (en) | Vehicle collision accident notification system and method | |
| CN109466488B (en) | Automatic vehicle collision rescue alarm system | |
| US20140132404A1 (en) | Pedestrian collision detection system, pedestrian collision notification system, and vehicle collision detection system | |
| CN106230940B (en) | A kind of vehicle collision detection method and system based on vehicle intelligent terminal | |
| CN104149729B (en) | The method and system of identification traveling vehicle periphery barrier | |
| CN105072413B (en) | A kind of intelligent travelling crane monitoring system and control method thereof based on DVR | |
| CN105319046A (en) | Method of collision detection and alarm and portable device of collision detection and alarm | |
| CN104442658B (en) | Automobile near-drowning life-saving system | |
| US11926319B2 (en) | Driving monitoring device and computer readable medium | |
| CN106920293A (en) | The automatic log analysis methodology of car accident | |
| KR101416528B1 (en) | Wireless smart system for vehicle safety monitoring and accident prevention | |
| CN105844964A (en) | Vehicle safe driving early warning method and device | |
| CN204821442U (en) | Vehicle SOS device and vehicle rescue system | |
| CN103544485B (en) | Driver recognition method and device based on intelligent terminal | |
| JP2021002112A (en) | Driving support device of motor cycle | |
| JP2016066231A (en) | Collision prevention device, collision prevention method, collision prevention program, and recording medium | |
| JP2015207049A (en) | Vehicle accident situation prediction device, vehicle accident situation prediction system and vehicle accident notification device | |
| CN110626259A (en) | Triangular warning frame and collision early warning method thereof | |
| US10298745B2 (en) | Method and apparatus for modifying interaction with a mobile-device | |
| KR20180136065A (en) | Parking impact detecting device performing notification of an impact of a vehicle and method of notifying an impact of a vehicle using a parking impact detecting device | |
| Lupinska-Dubicka et al. | In-car ecall device for automatic accident detection, passengers counting and alarming |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20200117 |
|
| AD01 | Patent right deemed abandoned |