CN105989721A - Lamp signal warning device and method thereof - Google Patents
Lamp signal warning device and method thereof Download PDFInfo
- Publication number
- CN105989721A CN105989721A CN201510073216.5A CN201510073216A CN105989721A CN 105989721 A CN105989721 A CN 105989721A CN 201510073216 A CN201510073216 A CN 201510073216A CN 105989721 A CN105989721 A CN 105989721A
- Authority
- CN
- China
- Prior art keywords
- cresset
- carrier
- road image
- traffic sign
- light
- 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
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000012545 processing Methods 0.000 claims abstract description 41
- 230000000284 resting effect Effects 0.000 claims description 44
- 230000000877 morphologic effect Effects 0.000 claims description 25
- 230000001143 conditioned effect Effects 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 10
- 238000012546 transfer Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract 2
- 230000003068 static effect Effects 0.000 abstract 2
- 230000003935 attention Effects 0.000 description 14
- 238000009434 installation Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000010332 selective attention Effects 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 3
- 230000002459 sustained effect Effects 0.000 description 3
- 102220588433 Keratin, type I cytoskeletal 18_S42A_mutation Human genes 0.000 description 2
- 102220536512 THAP domain-containing protein 1_S52A_mutation Human genes 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910044991 metal oxide Inorganic materials 0.000 description 2
- 150000004706 metal oxides Chemical class 0.000 description 2
- 102220102960 rs9895963 Human genes 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 102220497126 5-hydroxytryptamine receptor 3B_S51A_mutation Human genes 0.000 description 1
- 102220585558 T cell receptor gamma constant 1_S41A_mutation Human genes 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 239000002085 irritant Substances 0.000 description 1
- 231100000021 irritant Toxicity 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention provides a lamp signal warning device and a method thereof. The image capturing unit is used for continuously shooting road images of the vehicle. The processing unit is configured to execute a program comprising: a signal detection module, which sets a tracking frame for tracking the signal of the traffic signal corresponding to the traffic signal; the driving detection module is used for judging the driving state of the carrier so as to determine whether the carrier is in a static state; and a decision module, which starts a decision mode when the vehicle is judged to be in the static state and outputs a warning signal when the light signal of the traffic signal reaches a preset condition.
Description
Technical field
The present invention is related to a kind of cresset alarming device and method thereof, a kind of in order to detect livery state to point out cresset alarming device and the method thereof of driver.
Background technology
Attention belongs to a kind of psychology concept, local irritant level of consciousness will be caused to improve by controlling intelligence, and make concentrating in certain part things of selectivity of perception.In general it is noted that power can be divided into selective attention, Sustained attention power and dispersibility attention.Selective attention refers to the most whether can be primarily focused on the ability of single activity, and being susceptible to other external uncorrelated stimulation is affected；Sustained attention power refers to attention be maintained the ability in continuous print activity within a period of time；Dispersibility attention refers to carry out two pieces or the ability of above task in the same time.For drive for, drive need height absorbed power, the task of driving must have precedence over the task that other is secondary, therefore drive must strengthen particular for the part of selective attention and Sustained attention power.
Generally being driven in waiting the time that livery becomes lamp, attention is easily attracted by other secondary task, makes the priority layer of secondary task exceed the task of driving on the contrary.Now, driving easily because diverting one's attention to do other thing or the state of mind is not concentrated, causing livery but can not find when becoming lamp.
Generally solving the problem that driving demand power is not concentrated, part high-order vehicle is driven auxiliary machine by configuration and is assisted to promote the attention driven.Drive auxiliary machine and there is collection eyes, facial image and the function of vehicle inside and outside state, collected environmental data is analyzed, it is judged that the attention situation of driving and real-time vehicle condition, and actively record and prompting traveling state of vehicle.But, auxiliary machine of driving the most on the market all fails to assist to drive the situation following the trail of livery within the livery time that driving waits, makes still to be unknown to move ahead when driving in livery change lamp, in turn results in traffic problems.
Summary of the invention
Main purpose of the present invention, be solve drive because of divert one's attention and not it will be noted that livery become lamp problem.
For reaching above-mentioned purpose, the present invention provides a kind of cresset alarming device, includes an image acquisition unit, and a processing unit being connected to this image acquisition unit.This image acquisition unit is continuously shot the road image of a carrier.This processing unit is configured to perform following procedure, and this program includes: a livery detecting module, according to this road image, detects the cresset of a traffic sign；One driving detecting module, it is judged that the driving states of this carrier is to confirm whether this carrier is in a resting state；And a decision-making module, in judging that this carrier starts a decision-making mode when being in this resting state, and in the cresset of this traffic sign reach one pre-conditioned time, export an alerting.
Further, this livery detecting module according to position, color or morphological characteristic detect this road image may be for the composition of this traffic sign, and to setting the tracking framework of cresset of this traffic sign of tracking by traffic sign.
Further, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or confirms whether this carrier is in this resting state by this driving detecting module by the amount of change calculated in this road image.
Another object of the present invention is to provide a kind of cresset alarming device, includes an image acquisition unit, and a processing unit being connected to this image acquisition unit.This image acquisition unit is continuously shot the road image of carrier.This processing unit is configured to perform following procedure, and this program includes: a taillight detecting module, according to this road image, and the cresset of tail-light before detecting one；One driving detecting module, is that the driving states judging this carrier is to confirm whether this carrier is in a resting state；And a decision-making module, lie in and when this carrier of judgement is in this resting state, start a decision-making mode, and in the cresset of this front tail-light reach one pre-conditioned time, export an alerting.
Further, this taillight detecting module according to position, color or morphological characteristic detect this road image may be for the composition of this front tail-light, and to setting the tracking framework of cresset of this front tail-light of tracking by front truck taillight.
Further, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or calculates the amount of change in this road image by this driving detecting module and confirm whether this carrier is in this resting state.
Another object of the present invention is to provide a kind of cresset alarming device, includes an image acquisition unit, and a processing unit being connected to this image acquisition unit.This image acquisition unit is continuously shot the road image of carrier.This processing unit is configured to perform following procedure, and this program includes: a livery detecting module, according to this road image, detects the cresset of a traffic sign；One taillight detecting module, according to this road image, the cresset of tail-light before detecting one；One driving detecting module, is that the driving states judging this carrier is to confirm whether this carrier is in a resting state；And a decision-making module, in judging that this carrier starts a decision-making mode when being in this resting state, and in the cresset of this traffic sign or this front tail-light reach one pre-conditioned time, export an alerting.
Further, what this road image was detected in this livery detecting module foundation position, color or morphological characteristic may be the composition of this traffic sign, and to setting the tracking framework of cresset of this traffic sign of tracking by traffic sign, this taillight detecting module according to position, color or morphological characteristic detect this road image may be for the composition of this front tail-light, and to setting the tracking framework of cresset of this front tail-light of tracking by front truck taillight..
Further, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or calculates the amount of change in this road image by this driving detecting module and confirm whether this carrier is in this resting state..
Another object of the present invention is to provide a kind of cresset alarming method for power, includes: be continuously shot the road image of carrier；According to this road image, detect the cresset of a traffic sign；According to this road image, the cresset of tail-light before detecting one；Judge that the driving states of this carrier is to confirm whether this carrier is in a resting state；When this carrier is in this resting state, and in the cresset of this traffic sign or this front tail-light reach one pre-conditioned time, export an alerting.
Further, the step system of the cresset of this detecting traffic sign comprises: what this road image was detected in foundation position, color or morphological characteristic may be the composition of traffic sign, and to setting the first tracking framework of cresset of this traffic sign of tracking by traffic sign, this first is followed the trail of framework and is set according to following methods: define the region being likely to occur this traffic sign according to this road image；Hunting out in this region may be for the color lump of the color character of this traffic sign as fixed reference feature；This first scope following the trail of framework is determined according to the position of this color lump, ratio or by this connected region, color lump week side block edge.
Further, reach pre-conditioned described in and carry out decision-making according to following method: first follow the trail of framework by this and keep track the digital figure in the cresset of this traffic sign or this traffic sign；In this traffic sign cresset by red light transfer to green light or in the numerical value corresponding to this digital figure less than or equal to a default value time export this alerting.
Further, before this detecting, the step of the cresset of tail-light comprises: according to position, color or morphological characteristic detect this road image may be for the composition of this front tail-light, and to setting the second tracking framework of cresset of this front tail-light of tracking by front truck taillight；This second tracking framework is set according to following methods: define the region being likely to occur front vehicles according to this road image；Hunting out according to morphological characteristic in this region may be for the object of this front vehicles；The scope of this second tracking framework is determined according to position, ratio or the edge by this object of this object.
Further, reach pre-conditioned described in and carry out decision-making according to following method: keep track the cresset of this front tail-light by this second tracking framework；This alerting is exported when the brake light cresset of this front tail-light extinguishes.
Further, described resting state is identified according to following methods: is continuously shot the road image of this carrier or is linked to drive a vehicle computer to obtain each portion parameter of this carrier；Calculate the variable quantity of striograph frame before and after this road image, and when this variable quantity is less than a threshold value, it is determined that this carrier operates in this resting state；Or judge whether this carrier operates in this resting state by rotating speed, the braking parameters of carrier.
Therefore, the present invention has a following excellent effect compared to aforementioned known techniques:
1. the present invention is for following the trail of the livery in environment and front tail-light image, in livery by red turn green time, prompting driver that can be in good time notes, in order to driving in the event that attention is quickly transferred to drive.
2. the present invention can be integrated into driving recorder or drive on auxiliary machine, drives auxiliary function with other and uses, it is provided that drive perfect guiding instruction.
Accompanying drawing explanation
Fig. 1 is the block schematic diagram of cresset alarming device of the present invention.
Fig. 2 is the use view of cresset alarming device of the present invention.
Fig. 3 discloses captured road image.
Fig. 4 discloses traffic sign front image.
Fig. 5 discloses the image schematic diagram of traffic sign.
Fig. 6 discloses captured track schematic diagram.
Fig. 7 discloses front truck tailstock front image.
Fig. 8 discloses front truck tailstock image schematic diagram.
Fig. 9 discloses the schematic flow sheet () of cresset alarming method for power of the present invention.
Figure 10 discloses the schematic flow sheet (two) of cresset alarming method for power of the present invention.
Figure 11 discloses the schematic flow sheet (three) of cresset alarming method for power of the present invention.
Figure 12-A discloses the schematic flow sheet (four) of cresset alarming method for power of the present invention.
Figure 12-AA discloses the schematic flow sheet (five) of cresset alarming method for power of the present invention.
Figure 12-B discloses the schematic flow sheet (six) of cresset alarming method for power of the present invention.
Figure 13-A discloses the schematic flow sheet (seven) of cresset alarming method for power of the present invention.
Figure 13-B discloses the schematic flow sheet (eight) of cresset alarming method for power of the present invention.
Figure 13-C discloses the schematic flow sheet (nine) of cresset alarming method for power of the present invention.
Label declaration:
100 cresset alarming devices；
10
Image acquisition unit；
11
Road image；
20
Processing unit；
A1
Livery detecting module；
A2
Taillight detecting module；
A3
Driving detecting module；
A4
Decision-making module；
211A first follows the trail of framework；
211B first follows the trail of framework；
221A second follows the trail of framework；
221B second follows the trail of framework；
30
Storage element；
40
Alarm unit；
M
The position at cresset color lump center；
K1
With reference to constant；
K2
With reference to constant；
K3
With reference to constant；
K4
With reference to constant；
N
The central point of object；
S
Two border constants；
L
Leg of frame long constant；
D1
Boundary position；
D2
Boundary position；
FR
Second quadrant；
Step S1-step S5；
Step S21-step S23；
Step S31-step S33；
Step S41A-step S42A；
Step S421A-step S423A；
Step S41B-step S42B；
Step S51A-step S52A；
Step S51B-step S52B；
Step S51C-step S52C.
Detailed description of the invention
The relevant detailed description of the invention and technology contents, now just coordinate graphic being described as follows.Furthermore, graphic in the present invention, for convenience of explanation, its ratio may not be drawn to scale, and has the situation of exaggerating, and these graphic and ratios also are not used to limit the scope of the invention.
The present invention provides a kind of cresset alarming device and method thereof, is below described separately for multiple embodiments:
Cresset alarming device:
Refer to Fig. 1 and Fig. 2, for cresset alarming device of the present invention block schematic diagram and use view, as shown in the figure:
Cresset alarming device 100 of the present invention, consists predominantly of an image acquisition unit 10, and one is linked to the processing unit 20 of this image acquisition unit 10, and one in order to store or the storage element 30 of temporal data, and an alarm unit 40 being linked to this processing unit 20.
Described image acquisition unit 10 is arranged on carrier, and the road image 11 taken in travelling in order to be continuously shot this carrier, described road image 11 stores or is temporarily stored in this storage element 30.This image acquisition unit 10 can be any to have Charged Coupled Device (Charge coupled
Device, CCD) camera lens, complementary metal oxide semiconductors (CMOS) pipe (Complementary Metal Oxide
Semiconductor Transistors, CMOS) camera lens or the camera of infrared ray camera lens, can also be the image capture equipment that can obtain depth information, e.g. degree of depth camera (Depth
Or stereoscopic camera Camera).
Described processing unit 20 and this storage element 30 can collectively form computer or the processor of a computer or processor, e.g. personal computer, work station, host computer or other pattern, are not limiting as its kind at this.
In a preferred embodiment, this processing unit 20 is coupled to this storage element 30.This processing unit 20 can be central processing unit (Central
Processing Unit, CPU), or other programmable there is the microprocessor (Microprocessor) of general service or specific use, Digital System Processor (Digital Signal Processor, DSP), programmable controller, ASIC (Application Specific Integrated
Circuits, ASIC), programmable logic device (Programmable Logic Device, PLD) or other similar device or the combination of these devices.
Described storage element 30 can be the fixed or movable random access memory (Random of any kenel
Access Memory, RAM), read only memory (Read-Only
Memory, ROM), flash memory (Flash Memory) or similar assembly or the combination of said modules.This storage element 30 also can be made up of one or more accessible non-volatile memory organization packets.Specifically, it can be hard disk, memory card, also can be integrated circuit or firmware (firmware).
Described alarm unit 40 can be that speaker, cresset, display device or any other can point out warning message to the outut device of user (driver), is not intended to be limited the kind of this alarm unit 40 in the present invention.
Described processing unit 20 can load and perform following procedure, and described program is different according to its task, includes livery detecting module A1, taillight detecting module A2, driving detecting module A3 and decision-making module A4.Carrying out for livery detecting module A1, taillight detecting module A2, driving detecting module A3 and decision-making module A4 and processing individually below of task illustrates respectively:
Livery detecting module A1
Described livery detecting module A1 is in order to according to this road image, to detect the cresset of a traffic sign.What this road image 11 was detected in this livery detecting module A1 foundation position, color or morphological characteristic may be the composition of this traffic sign, and to first tracking framework 211A, 211B (refer to Fig. 4 and Fig. 5) of cresset of this traffic sign of tracking should be set by traffic sign, use the real-time cresset following the trail of this traffic sign.
Disclosed below a kind of feasible it is embodied as aspect, this livery detecting module A1 is via this storage element 30(or directly via this image acquisition unit 10) obtain the most several road images 11, and carry out image processing by image processing module for this road image 11, image is adjusted to the form being prone to this livery detecting module A1 identification.
See also shown in Fig. 3, taken road image 11 is defined as four quadrants by this livery detecting module A1, and the second quadrant FR(therein is defaulted as a left side drives, if the right side is driven, set first quartile) area-of-interest that is set as in image, and search any object meeting corresponding livery feature for this area-of-interest.Described livery feature can be by the data base set up for different livery forms in advance, and in time starting detecting, this livery detecting module A1 judges the type belonging to livery according to color or the morphological characteristic of livery and positions this livery coordinate in image.
After obtaining described coordinate, this livery detecting module A1 determines this first scope following the trail of framework 211A, 211B according to the position of this color lump, ratio or by this connected region, color lump week side block edge.
Hereinafter it is aided with graphic illustrating for two different embodiments:
(1) by connection block definition the first tracking framework 211A
Refer to Fig. 4, preferably implement in aspect in one, the position of this livery cresset color lump can be set as center by this livery detecting module A1, and detect the connection block of this color lump week side, the outline of this connection block is directly set as this first tracking framework 211A, in order to follow the trail of all cressets in this livery.
By this first follow the trail of framework 211A, this livery detecting module A1 can real-time tracing this first follow the trail of all cressets in framework 211A, the state current to judge this cresset.
(2) by cresset ratio definition the first tracking framework 211B
Referring to Fig. 5, preferably implement in aspect in one, the position of this livery cresset color lump can be set as center by this livery detecting module A1, and arranges one first tracking framework 211B according to the size of this cresset color lump.As shown in the figure, first length direction is set in taken image as x, width is y, set the position at this cresset color lump center as M (x1, y1), continue and calculate the area of this color lump, girth or diameter, by this area, girth or diameter obtain, according to ratio set in advance, the color lump area that at least four such as calculates with reference to constant K1-K4(and account for 300 pixel, the pixel tried to achieve occupied by length and width according to livery ratio is respectively 30, 10 pixel, and after 2, respectively it is multiplied by 1.2 times of acquirements four with reference to constant K1-K4 as the first border following the trail of framework 211B as basic numerical value).Finally by this reference constant K1-K4 define four border E (x,
y;X1+K1 > x > x1-K2, y1+K3 > y > y1-K4), by the first tracking framework 211B described in the region definition that these four borders are defined.
After setting this first tracking framework 211B, this livery detecting module A1 will can note down corresponding data according to the kind of cresset, described data such as livery kind (traffic lights, level crossing livery, direction livery, numeral livery, pedestrians travel's livery etc.), different by acquired parameter, will make different decision-makings, this part will have further description in the decision-making module A4 of back segment.
Taillight detecting module A2
This taillight detecting module A2, in order to according to this road image, detects the cresset of tail-light before.This taillight detecting module A2 according to position, color or morphological characteristic detect this road image 11 may be for the composition of this front tail-light, and to setting second tracking framework 221A, 221B of cresset of this front tail-light of tracking by front truck taillight.
Disclosed below a kind of feasible it is embodied as aspect, this taillight detecting module A2 is via this storage element 30(or directly via this image acquisition unit 10) obtain the most several road images 11, and carry out image processing by image processing module for this road image 11, image is adjusted to the form being prone to this taillight detecting module A2 identification.
See also shown in Fig. 6, taken road image 11 is set a track detecting block by this taillight detecting module A2, define the region being likely to occur front vehicles, detecting block in this track can determine scope by the ground extension line in track, or determined by set in advance two asymptotes (based on the perspective concept that image is far and near) being divided into two sides, left and right, the area-of-interest that this track detecting block is set as in image, and search any object meeting corresponding taillight feature for this area-of-interest.Described taillight feature can be by the data base set up for different vehicle form in advance, and in time starting detecting, this taillight detecting module A2 judges according to the morphological characteristic of general vehicle may be as the object of vehicle.
Obtain described may be vehicle object after, this taillight detecting module A2 according to position, ratio or the edge by this object of this object determine this second follow the trail of framework 221A, 221B scope.
(1) by object scale definition the second tracking framework 221A
Refer to Fig. 7, preferably implement in aspect in one, first set in taken image length direction as x, width is y, and after this taillight detecting module A2 searches this object meeting corresponding front vehicles by morphological characteristic, the central point according to this object of weight setting is N (x2, y2), continue the area or path length calculated in image occupied by this object entirety, and according to ratio set in advance, this area or path length are obtained two border constant S, and leg of frame long constant L.Substitute into such proportionality constant to this central point N (x2, y2) use and calculate the boundary position D1 (x2+S corresponding to this object two side, y2), D2 (x2-S, y2), continue and this boundary position D1, D2 are set as center again and substitute into this leg of frame long constant L as boundary value D1 (x, y; x2 + S + L/2 > x > x2 + S - L/2, y2 + L/2
> y > y2 - L/2 )、D2( x, y; x2 - S
+ L/2 > x > x2-S-L/2, y2+L/2 > y > y2-L/2) to define the second described tracking framework 221A.
(2) framework 221B is followed the trail of by the Edge definition second of this object
Refer to Fig. 8, preferably implement in aspect in one, after this taillight detecting module A2 searches this object meeting corresponding front vehicles by morphological characteristic, weight proportion find the position of two sides in this object.Continue, by this two side position according to color or the difference of foundation GTG degree, find this object meets taillight feature object (such as red block, yellow block or with the week side visibly different block of connected region), capture the border of this taillight, use and follow the trail of framework 221B as second.
After setting this second tracking framework 221B, according to the change of cresset, this taillight detecting module A2 will can judge whether front truck has started to travel, such as judge whether front truck travels or judge whether front truck sails out of by front truck cresset position in image or area (the most remote area is the least) by the variable quantity of brightness before and after the cresset of front truck, and these part data are sent to this decision-making module A4 as reference data.
Above-mentioned livery detecting module A1 and taillight detecting module A2 can not limit in the present invention by respectively or same image processing module or independent image processing chip process for the program of image processing.Described image processing module is primarily useful for carrying out such as GTG process, binary conversion treatment, filter process, colouring process, feature extraction, image subtract each other or this image processing program of other class, is not limited in the present invention.
Driving detecting module A3
This driving detecting module A3 judges that the driving states of this carrier is to confirm whether this carrier is in a resting state.Two kinds of feasible enforcement aspects disclosed below:
(1) judge by driving computer
The each portion parameter reading carrier used by the driving computer that described processing unit 20 can be linked on carrier.Specifically, described driving computer can be the engine controller (Engine monitoring vehicle-state
Control Unit, ECU) or any other may determine that the monitoring device of driving states, is not intended to be limited the kind of this driving computer in the present invention.When vehicle launch, this driving detecting module A3 actively captures the parameter that this driving computer is recorded, and thereby, can determine whether that carrier is in traveling or halted state.The parameter captured can be such as the gear that the speed revolution value of engine, brake numerical value or vehicle are current, is not limited in the present invention.
(2) judge by front and back's image variable quantity
Described driving detecting module A3 can be directly accessed the road image 11 in this storage element 30, and calculates the variable quantity of striograph frame before and after this road image 11, and when this variable quantity is less than a threshold value, it is determined that this carrier operates in a resting state.Specifically, this driving detecting module A3 first can carry out GTG process by the figure of road image 11 is auspicious before and after this respectively, schemes auspicious grey-tone image before and after acquirement.Continue and two grey-tone images are subtracted each other by image subtractive method, the variation scattergram of auspicious image is schemed before and after acquirement, the most again by this variation scattergram via binary conversion treatment, obtain the pixel variable quantity of variation, judge whether vehicle is in the state of traveling by the pixel quantity of change, when described variable quantity is less than a threshold value, it is determined that this carrier operates in this resting state.But, the above-mentioned mode only present invention wherein one can be embodied as aspect for reference, and the present invention is not intended to be limited to above-mentioned enforcement aspect.
Decision-making module A4
This decision-making module A4, after obtaining above-mentioned parameter and variable, will produce the different results of decision based on acquired parameter is different.This decision-making module A4 starts a decision-making mode when detecting resting state, and in the cresset of this traffic sign reach one pre-conditioned time, export an alerting.Described pre-conditioned and corresponding warning message is as follows:
Situation one, only detect the cresset of traffic sign
When front is without vehicle, drive and stop because of red light, now this driving detecting module A3 judges that carrier is in a resting state, in this resting state, this decision-making module A4 starts a decision-making mode, in this livery detecting module A1 by this first follow the trail of framework 211A, 211B detect this red light become green light time, described decision-making module A4 informs drive the warning message (being applicable to general traffic lights) that changed of cresset by transmitting one by this alarm unit 40.In another preferred embodiment, when this livery detecting module A1 by this first follow the trail of framework 211A, 211B follow the trail of on this cresset the threshold value that shown numeral is preset less than one time, described decision-making module A4 informs to drive become the warning message (being applicable to the livery with cresset of countdowning) of lamp remaining time by transmitting one by this alarm unit 40.In another preferred embodiment, when this livery detecting module A1 by this first follow the trail of framework 211A, 211B follow the trail of show a turning pattern on this cresset time, this decision-making module A4, according to the direction of turning pattern, transmits one by this alarm unit 40 and informs the left curved or right curved warning message (being applicable to the livery with blinker number) of driving.
Situation two, only detect front taillight cresset
When there is vehicle in front, but fails to photograph livery (such as front vehicles is covered livery or is Ordinary Rd traffic congestion), now this driving detecting module A3 judges that carrier is in a resting state, in this resting state, this decision-making module A4 starts a decision-making mode, when this taillight detecting module A2 detects the taillight cresset change of front vehicles by this second tracking framework 221A, 221B, described decision-making module A4 informs, by transmitting one by this alarm unit 40, the warning message that driving front truck vehicle has travelled.
Situation three, detect traffic sign and the cresset of front taillight simultaneously
Driving carrier to stop because of red light, now this driving detecting module A3 judges that carrier is in a resting state.In this resting state, this decision-making module A4 starts a decision-making mode, and obtains livery and the state of front truck by this livery detecting module A1 and taillight detecting module A2.In this decision-making mode, when being simultaneously detected such as preceding vehicle tail lights and front livery, except the situation described in above-mentioned situation one and situation two, this decision-making module A4 can carry out decision-making by the state of the state of livery cresset and front tail-light cresset, such as when livery by red light transfer to the precondition of green light reach time, by this taillight detecting module A2, this decision-making module A4 judges whether the cresset of front tail-light changes, when the cresset being detected this front tail-light changes, then transmit a warning message informing driving demand by this alarm unit 40；Current vehicle taillight is not reminded when not changing.This aspect is applicable to situation during road traffic congestion, and can reduce the probability of erroneous judgement.
More than have been for hardware of the present invention part and the task of corresponding execution be described in detail, continue, will be further illustrated for the inventive method step.
Cresset alarming method for power
See also Fig. 9, be the present invention first schematic flow sheet of implementing aspect, as shown in the figure:
The present invention provides a cresset alarming method for power, and the method can coordinate electronic installation to use.This electronic installation can be but be not limited to a driving recorder, drives this electronic installation of auxiliary machine, aforementioned cresset alarming device 100 etc. or other class.This electronic installation includes image acquisition unit 10, is linked to the processing unit 20 of this image acquisition unit 10, in order to store or the storage element 30 of temporal data, and is linked to the alarm unit 40 of this processing unit 20.
Described cresset alarming method for power includes below step:
First, when this carrier stops, the road image 11 of carrier it is continuously shot by image acquisition unit 10；(step S1)
Now, this processing unit 20, according to this road image 11, detects the cresset of a traffic sign.Specifically, this processing unit 20 is detected in this road image 11 according to position, color or morphological characteristic may be for the composition of traffic sign, and to setting first tracking framework 211A, 211B of cresset of this traffic sign of tracking by traffic sign；(step S2)
While following the trail of this traffic sign, this processing unit 20, according to this road image 11, detects the cresset of tail-light before.Specifically, this processing unit 20 according to position, color or morphological characteristic detect this road image 11 may be for the composition of front tail-light, and to setting second tracking framework 221A, 221B of cresset of this traffic sign of tracking by front truck taillight；(step S3)
Judge that the driving states of this carrier is to confirm whether this carrier is in a resting state (this step can be implemented in step S2 and step S3 previous crops as trigger condition, also may be disposed at after step S5 as final decision program, is not intended to be limited in the present invention)；(step S4)
When judging that this carrier is in this resting state, and the cresset of this traffic sign or this front tail-light meet one pre-conditioned time, export an alerting to this alarm unit 40, provide by this alarm unit 40 and drive correct instruction.(step S5)
Seeing also Figure 10, in step S2, this first tracking framework 211A is set according to following methods:
The region being likely to occur this traffic sign is defined according to this road image 11.In this step, by this livery detecting module A1, taken road image 11 can be defined as four quadrants, and if the second quadrant FR(right side therein is driven, set first quartile) area-of-interest that is set as in image；(step S21)
Hunting out in this region may be for the color lump of the color character of this traffic sign as fixed reference feature.While following the trail of this traffic sign by this color lump, the object of traffic sign can be met by morphological characteristic search, to reduce the probability of identification mistake；(step S22)
This first scope following the trail of framework 211A is determined according to the position of this color lump, ratio or by this connected region, color lump week side block edge.(step S23)
Seeing also Figure 11, in step S3, this second tracking framework 221A, 221B are set according to following methods:
The region being likely to occur front vehicles is defined according to this road image 11.Scope can be determined by the ground extension line in track, or be determined by set in advance two asymptotes (based on the perspective concept that image is far and near) being divided into two sides, left and right, the area-of-interest that this track detecting block is set as in image；(step S31)
Hunting out according to morphological characteristic in this region may be for the object of this front vehicles.Can be by the data base set up for different vehicle form in advance, in time starting detecting, utilizing Adaboost algorithm to judge according to the morphological characteristic of general vehicle through this taillight detecting module A2 may be as the object of vehicle；(step S32)
The scope of this second tracking framework 221A, 221B is determined according to position, ratio or the edge by this object of this object.(step 33)
See also Figure 12-A and Figure 12-B, in step S4, described resting state is identified according to following two kinds of methods:
Method one, (such as Figure 12-A)
It is continuously shot the road image 11 of carrier；(step S41A)
Calculate the variable quantity of this road image 11 striograph frame front and back, and when this variable quantity is less than a threshold value, it is determined that this carrier operates in a resting state.(step S42A)
Referring to Figure 12-AA, preferably implement in aspect in one, step 42A can be analyzed to below step:
Carry out GTG process the most respectively by the figure of road image 11 is auspicious before and after this, before and after acquirement, scheme auspicious grey-tone image；(step S421A)
Continue and two grey-tone images are subtracted each other by image subtractive method, before and after acquirement, scheme the variation scattergram of auspicious image；(step S422A)
Again by this variation scattergram via binary conversion treatment, obtain the pixel variable quantity of variation；(step S423A)
When described variable quantity is less than a threshold value, it is determined that this carrier operates in this resting state.(step S424A)
Method two, (such as Figure 12-B)
It is linked to drive a vehicle computer to obtain each portion parameter of this carrier；(step S41B)
A resting state whether is operated in by rotating speed, this carrier of car parameter decision of carrier.(step S42B)
See also " Figure 13-A " to " Figure 13-C ", in step S5, described in reach pre-conditioned system and carry out decision-making according to three kinds of following methods:
Method one, (Figure 13-A)
Follow the trail of framework 211A, 211B by first and keep track the cresset of this traffic sign；(step S51A)
This alerting is exported when the cresset of this traffic sign is transferred to green light by red light.(step S52A)
Method two, (Figure 13-B)
Follow the trail of framework 211A, 211B by first and search the digital figure in this traffic sign；(step S51B)
Note down this digital figure, and export this alerting when the numerical value corresponding to this digital figure is less than or equal to a default value.(step S52B)
Method three, (Figure 13-C)
Follow the trail of framework 221A, 221B by second and keep track the cresset of this front tail-light；(step S51C)
This alerting is exported when the brake light cresset of this front tail-light extinguishes.(step S52C)
The inventive method step can act also as a kind of software program and implements, in order to be stored in the computer-readable medium storings such as disc, hard disk, semiconductor memory, and it is placed on electronic installation by this computer-readable medium storing and is accessed use by this electronic installation.Specifically, this electronic installation can be small-sized portable electronic devices, recording journey device, driving safety auxiliary device, the ODB(On-board diagnostics of Vehicular system), this electronic installation etc. of computer equipment or class.
In sum, the present invention for following the trail of the livery in environment and front tail-light image, in livery by red turn green time, prompting driver that can be in good time notes, in order to driving in the event that attention is quickly transferred to drive.The present invention can be integrated into driving recorder or drive on auxiliary machine, drives auxiliary function with other and uses, it is provided that drive perfect guiding instruction.
Below the present invention is described in detail, but the above, only a preferred embodiment of the present invention, when not limiting the scope of the present invention with this, the most all impartial changes made according to claims of the present invention and modification, all should still belong in patent covering scope of the present invention.
Claims (15)
1. a cresset alarming device, comprises:
One image acquisition unit, is continuously shot the road image of a carrier；
One processing unit, is configured to perform following procedure, and this program includes:
One livery detecting module, according to this road image, detects the cresset of a traffic sign；
One driving detecting module, it is judged that the driving states of this carrier is to confirm whether this carrier is in a resting state；And
One decision-making module, starts a decision-making mode when this carrier is in this resting state in judging, and in the cresset of this traffic sign reach one pre-conditioned time, export an alerting.
2. cresset alarming device as claimed in claim 1, it is characterized in that, this livery detecting module detects in this road image the composition for this traffic sign according to position, color or morphological characteristic, and to setting the tracking framework of cresset of this traffic sign of tracking by traffic sign.
3. the cresset alarming device as described in any one of claim 1 to 2, it is characterized in that, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or confirms whether this carrier is in this resting state by this driving detecting module by the amount of change calculated in this road image.
4. a cresset alarming device, it is characterised in that comprise:
One image acquisition unit, is continuously shot the road image of carrier；
One processing unit, is configured to perform following procedure, and this program includes:
One taillight detecting module, according to this road image, the cresset of tail-light before detecting one；
One driving detecting module, it is judged that the driving states of this carrier is to confirm whether this carrier is in a resting state；And
One decision-making module, starts a decision-making mode when this carrier is in this resting state in judging, and in the cresset of this front tail-light reach one pre-conditioned time, export an alerting.
5. cresset alarming device as claimed in claim 4, it is characterized in that, this taillight detecting module detects in this road image the composition for this front tail-light according to position, color or morphological characteristic, and to setting the tracking framework of cresset of this front tail-light of tracking by front truck taillight.
6. the cresset alarming device as described in any one of claim 4 to 5, it is characterized in that, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or calculates the amount of change in this road image by this driving detecting module and confirm whether this carrier is in this resting state.
7. a cresset alarming device, it is characterised in that comprise:
One image acquisition unit, is continuously shot the road image of carrier；
One processing unit, is configured to perform following procedure, and this program includes:
One livery detecting module, according to this road image, detects the cresset of a traffic sign；
One taillight detecting module, according to this road image, the cresset of tail-light before detecting one；
One driving detecting module, it is judged that the driving states of this carrier is to confirm whether this carrier is in a resting state；And
One decision-making module, starts a decision-making mode when this carrier is in this resting state in judging, and in the cresset of this traffic sign or this front tail-light reach one pre-conditioned time, export an alerting.
8. cresset alarming device as claimed in claim 7, it is characterized in that, the composition in this road image for this traffic sign is detected in this livery detecting module foundation position, color or morphological characteristic, and to setting the tracking framework of cresset of this traffic sign of tracking by traffic sign, this taillight detecting module detects in this road image the composition for this front tail-light according to position, color or morphological characteristic, and to setting the tracking framework of cresset of this front tail-light of tracking by front truck taillight.
9. the cresset alarming device as described in any one of claim 7 to 8, it is characterized in that, this processing unit is linked to the driving computer of this carrier to obtain the driving states of this carrier, or confirms whether this carrier is in this resting state by this driving detecting module by the amount of change calculated in this road image.
10. a cresset alarming method for power, includes:
It is continuously shot the road image of carrier；
According to this road image, detect the cresset of a traffic sign；
According to this road image, the cresset of tail-light before detecting one；
Judge that the driving states of this carrier is to confirm whether this carrier is in a resting state；
When this carrier is in this resting state, and in the cresset of this traffic sign or this front tail-light reach one pre-conditioned time, export an alerting.
11. cresset alarming method for power as claimed in claim 10, it is characterized in that, the step of the cresset of this detecting traffic sign comprises: detect in this road image the composition for this traffic sign according to position, color or morphological characteristic, and to setting the first tracking framework of cresset of this traffic sign of tracking by traffic sign；
This first tracking framework is set according to following methods:
The region being likely to occur this traffic sign is defined according to this road image；
Hunting out in this region may be for the color lump of the color character of this traffic sign as fixed reference feature；
This first scope following the trail of framework is determined according to the position of this color lump, ratio or by this connected region, color lump week side block edge.
12. cresset alarming method for power as claimed in claim 11, it is characterised in that described in reach pre-conditioned and carry out decision-making according to following method:
Follow the trail of framework by first and keep track cresset or the digital figure of this traffic sign；
In this traffic sign cresset by red light transfer to green light or in the numerical value corresponding to this digital figure less than or equal to a default value time export this alerting.
The 13. cresset alarming method for power as described in claim 10, it is characterized in that, before this detecting, the step of the cresset of tail-light comprises: detecting in this road image according to position, color or morphological characteristic may be for the composition of this front tail-light, and to setting the second tracking framework of cresset of this front tail-light of tracking by front truck taillight；
This second tracking framework is set according to following methods:
The region being likely to occur front vehicles is defined according to this road image；
Hunting out according to morphological characteristic in this region may be for the object of this front vehicles；
The scope of this second tracking framework is determined according to position, ratio or the edge by this object of this object.
14. cresset alarming method for power as claimed in claim 13, it is characterised in that described in reach pre-conditioned and carry out decision-making according to following method:
The cresset of this front tail-light is kept track by this second tracking framework；
This alerting is exported when the brake light cresset of this front tail-light extinguishes.
15. cresset alarming method for power as claimed in claim 10, it is characterised in that described resting state is identified according to following methods:
It is continuously shot the road image of this carrier or is linked to drive a vehicle computer to obtain each portion parameter of this carrier；
Calculate the variable quantity of striograph frame before and after this road image, and judge that this carrier operates in this resting state when this variable quantity is less than a threshold value, or judge whether this carrier operates in this resting state by rotating speed, the braking parameters of carrier.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW103145454A TW201623067A (en) | 2014-12-25 | 2014-12-25 | Signal alarm device, method, computer readable media, and computer program product |
| TW103145454 | 2014-12-25 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105989721A true CN105989721A (en) | 2016-10-05 |
Family
ID=56984517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510073216.5A Pending CN105989721A (en) | 2014-12-25 | 2015-02-12 | Lamp signal warning device and method thereof |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105989721A (en) |
| TW (1) | TW201623067A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106971581A (en) * | 2017-04-26 | 2017-07-21 | 维沃移动通信有限公司 | A kind of traffic signal light condition based reminding method and mobile terminal |
| CN107749184A (en) * | 2017-11-03 | 2018-03-02 | 英属开曼群岛商麦迪创科技股份有限公司 | Traffic sign cresset changes system for prompting and method |
| CN109466446A (en) * | 2018-11-30 | 2019-03-15 | 南昌与德软件技术有限公司 | A kind of driving reminding method, device, automobile and storage medium |
| CN110660256A (en) * | 2019-10-22 | 2020-01-07 | 北京地平线机器人技术研发有限公司 | Method and device for estimating state of signal lamp |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI782971B (en) * | 2017-11-03 | 2022-11-11 | 英屬開曼群島商麥迪創科技股份有限公司 | Traffic light change reminding system and method thereof |
| CN111775944B (en) * | 2019-04-04 | 2021-12-28 | 富泰华工业（深圳）有限公司 | Driving assistance apparatus, method, and computer-readable storage medium |
-
2014
- 2014-12-25 TW TW103145454A patent/TW201623067A/en unknown
-
2015
- 2015-02-12 CN CN201510073216.5A patent/CN105989721A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106971581A (en) * | 2017-04-26 | 2017-07-21 | 维沃移动通信有限公司 | A kind of traffic signal light condition based reminding method and mobile terminal |
| CN107749184A (en) * | 2017-11-03 | 2018-03-02 | 英属开曼群岛商麦迪创科技股份有限公司 | Traffic sign cresset changes system for prompting and method |
| CN109466446A (en) * | 2018-11-30 | 2019-03-15 | 南昌与德软件技术有限公司 | A kind of driving reminding method, device, automobile and storage medium |
| CN110660256A (en) * | 2019-10-22 | 2020-01-07 | 北京地平线机器人技术研发有限公司 | Method and device for estimating state of signal lamp |
| CN110660256B (en) * | 2019-10-22 | 2021-04-27 | 北京地平线机器人技术研发有限公司 | Method and device for estimating state of signal lamp |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201623067A (en) | 2016-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105989721A (en) | Lamp signal warning device and method thereof | |
| US11318928B2 (en) | Vehicular automated parking system | |
| US20180211121A1 (en) | Detecting Vehicles In Low Light Conditions | |
| CN104537841B (en) | Unlicensed vehicle violation detection method and detection system thereof | |
| US20150161796A1 (en) | Method and device for recognizing pedestrian and vehicle supporting the same | |
| CN104318206B (en) | A kind of obstacle detection method and device | |
| CN108847032A (en) | A kind of traffic violation recognition methods and device | |
| CN107054466A (en) | Parking assistance system and its application method for vehicle | |
| US20140028873A1 (en) | Image processing apparatus | |
| CN104537360A (en) | Method and system for detecting vehicle violation of not giving way | |
| CN107045619A (en) | Exterior environment recognition device | |
| US20180239972A1 (en) | Vehicle vision system with enhanced traffic sign recognition | |
| CN111891061B (en) | Vehicle collision detection method and device and computer equipment | |
| US11288569B2 (en) | Vehicle driving assist system with enhanced data processing | |
| JP4807354B2 (en) | Vehicle detection device, vehicle detection system, and vehicle detection method | |
| US20160034769A1 (en) | Vehicle vision system with traffic sign recognition | |
| CN104239847B (en) | Driving warning method and electronic device for vehicle | |
| US20170032198A1 (en) | Vehicle vision system with object detection | |
| JP6756908B2 (en) | Methods and devices for detecting pedestrians around the vehicle | |
| US20210303885A1 (en) | Generative adversarial network models for small roadway object detection | |
| CN107871398A (en) | A kind of method and system that traffic lights identification is carried out by drive recorder | |
| US20190325232A1 (en) | Two-wheel vehicle riding person number determination method, two-wheel vehicle riding person number determination system, two-wheel vehicle riding person number determination apparatus, and program | |
| CN111105619A (en) | Method and device for judging road side reverse parking | |
| KR102587047B1 (en) | Traffic Violation Enforcement System for Two-Wheeled Cehicle Which Supports Dual Operation Mode | |
| CN105389991A (en) | Self-adaptive snapshot method for behavior of running red light |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161005 |
|
| WD01 | Invention patent application deemed withdrawn after publication |