CN101934771A - Vehicle collision warning system - Google Patents

Vehicle collision warning system Download PDF

Info

Publication number
CN101934771A
CN101934771A CN2009101422832A CN200910142283A CN101934771A CN 101934771 A CN101934771 A CN 101934771A CN 2009101422832 A CN2009101422832 A CN 2009101422832A CN 200910142283 A CN200910142283 A CN 200910142283A CN 101934771 A CN101934771 A CN 101934771A
Authority
CN
China
Prior art keywords
image
direct vision
vehicle
obstacle
amount
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.)
Granted
Application number
CN2009101422832A
Other languages
Chinese (zh)
Other versions
CN101934771B (en
Inventor
陈加增
廖育升
高祥钧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Automotive Research and Testing Center
Original Assignee
Automotive Research and Testing Center
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Automotive Research and Testing Center filed Critical Automotive Research and Testing Center
Priority to CN200910142283A priority Critical patent/CN101934771B/en
Publication of CN101934771A publication Critical patent/CN101934771A/en
Application granted granted Critical
Publication of CN101934771B publication Critical patent/CN101934771B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses a vehicle collision warning system, which acquires an image of an indirect visual area in vehicle driving by at least one camera, continuously transmits the image of the indirect visual area into a processing unit and judges at least one barrier from the image, wherein the processing unit can calculate the travel speed and transverse movement of the barrier according to the change of the position of the barrier with time, presets at least one pre-warning area in the image, judges whether the barrier is to appear in the pre-warning area or not, and timely generates a pre-warning signal used for starting at least one warning indicator to produce sounds, light sources or the images to warn a driver to heighten vigilance. The vehicle collision warning system can be matched with a vehicle speed sensor, a brake system and a turn light system of a vehicle to perform leveled pre-warning according to danger levels to actively or passively inform the drivers of dangers, and has different information effects and responses.

Description

Vehicle collision alarming system
Technical field
The present invention is relevant with a kind of obstacle detection system, particularly about a kind of full size vehicle that is arranged on, in order to the obstacle in the non-direct vision of the monitoring driving person zone, and makes the vehicle collision alarming system that early warning is notified.
Background technology
Because Oversize vehicle car body height, vehicle body are long, cause the direct visual range of steerman to be vulnerable to obstruction and to become narrow, such as the place ahead of Oversize vehicle and about two sides all have many blind areas, and the main cause that described blind area often just causes the Oversize vehicle accident to keep pouring in, wherein especially with front bump to pedestrian or injury maximum that vehicle was caused, so Oversize vehicle the place ahead generally is provided with according to the ground mirror to avoid accidents.
Moreover, mention Oversize vehicle driver's non-direct vision position among the ECER46 of European Union, and the related specifications how forward and backward visor person that comes the driver assistance is set.But, observation mirror belongs to a kind of curved reflector, its observe car body rear and about during the mobile object of two sides, when judging with rule-of-thumb of drivers, the distance of conjecture car body and mobile object, the observational technique that belongs to passive and non-intuition, moreover, the setting of mirror body is layer by layer not only observed difficult, still the visual zone that exists easy carelessness to discover, and back mirror unavoidably belongs to passive alarm set, and can't reach the effect of initiatively reminding alarm, therefore, the non-direct vision district in Oversize vehicle car body the place ahead still must note just avoiding unexpected by steerman by every means and take place, therefore, and the generation of the accident that can't effectively prevent an accident.
In addition, in some prior arts, also there is the people to propose terms of settlement at the problems referred to above.Similarly be the novel patent M303843 in Taiwan number, it is mentioned and utilizes the wide-angle photography structure detection to drive the blind area zone that can't see, is arranged on operator's compartment with a read out instrument again and detects content to show; But, exist though utilize image can eliminate blind area zone, the detection picture that continues becomes the burden of steerman also easily, steerman and can't observation picture all the time in content, if divert one's attention to observe, also might cause unexpected the generation on the contrary.Moreover, the patent EP1988488 of European Union number blind spot region moving Object Detection system, use video camera equally and gather car body periphery picture, then will obtain picture segmentation again becomes several blocks, mat is judged the movability of cutting apart block, whether there is obstacle to detect the car body periphery, but this technology must be a plurality of blocks with picture segmentation, greatly then identification is poor when dividing block, easily cause erroneous judgement, though and the zoning is little can obtain higher identification, the relative sizable burden of treater that easily causes, easily over-burden that cause can't prompt alarm because of treater, the outer consequence that takes place of giving one's regards.
In view of this, the present invention proposes the high vehicle collision alarming system of a kind of image processing efficient then, especially at Oversize vehicle the place ahead and about the non-direct vision zone of two sides detect, and classification makes the early warning notice, can improve those disappearances that are present in the prior art in view of the above.
Summary of the invention
Main purpose of the present invention is providing a kind of vehicle collision alarming system, actv. utilizes image processing technique that careful and careful detection is carried out in the non-direct vision zone of Oversize vehicle according to this, and when dangerous property exists, make the early warning notice, effectively reduce the generation of contingency.
Of the present invention time a purpose is providing a kind of vehicle collision alarming system, obstacle hazard level according to the non-direct vision of Oversize vehicle driver district is carried out grade distinguishing, to select image, light source or sound as alarm signal, note initiatively to cause steerman according to each grade decision.
A further object of the present invention is providing a kind of vehicle collision alarming system, further with the existing of image complexity calculation mode disturbance in judgement thing, moving velocity and side travel amount, with erroneous judgement of reduction system or sensitive excessively probability.
Another purpose of the present invention sees through efficient image processing mode to judge the obstacle that whether has certain moving velocity of tool and tool side travel amount in the pre-police region in the place ahead for a kind of vehicle collision alarming system is provided, and then proposes caution when obstacle exists.
For achieving the above object, vehicle collision alarming system of the present invention, device image harvester on Oversize vehicle, processing unit and annunciator, utilize the image collection device to detect the non-direct vision zone of this Oversize vehicle and obtain at least one non-direct vision district image, with processing unit non-direct vision district image is carried out image processing again, tell the obstacle in the non-direct vision district image, and utilize obstacle position relative time variable quantity, estimate the moving velocity and the side travel amount of obstacle, moreover, processing unit is preset with a pre-police region in the non-direct vision zone position of Oversize vehicle, suppose that obstacle enters in the pre-police region and through judgement and has the side travel amount, processing unit will send an alarm signal; Annunciator electrically connects processing unit, and sends light source, image or sound prompting steerman behind the reception alarm signal and note contingent danger.
On the other hand, processing unit can also be done the obstacle of obtaining further to control to combine with the original speed of a motor vehicle perceptron of Oversize vehicle, brake system, turn signal system, sends different hazard level alarm signals according to the level difference.
Beneficial effect of the present invention is, detect and under image processing technique complements each other by image, as long as have mobile object in the non-direct vision district image, all can confirm to make and it will be set at the obstacle that to follow the trail of via data processing technique, moreover, the present invention is more in conjunction with the speed sensor of Oversize vehicle, system is considering that the Oversize vehicle quality is heavy, the factor that braking distance is long, when carrying out the subregion sensing according to speed signal, in addition, the present invention is again in conjunction with the brake system and the turn signal system of Oversize vehicle, for doing the hazard level classification in the position of obstacle, initiatively under critical situation, utilize sound and light program to cause that synchronously steerman notes, and under general situation, then take care in the zone of reminding steerman to be careful with picture or light source.
Beneath by the appended graphic explanation in detail of specific embodiment cooperation, when the effect that is easier to understand purpose of the present invention, technology contents, characteristics and is reached.
Description of drawings
Fig. 1 is the schematic top plan view of one embodiment of the invention;
Fig. 2 is the front elevational schematic of Fig. 1;
Fig. 3 is a structural representation of the present invention;
Fig. 4 is the present invention's first annunciator and the second annunciator scheme drawing;
Fig. 5 supposes the scheme drawing of an obstacle by the first non-direct vision district image for the present invention;
Fig. 6 is the place ahead of the present invention barrier judgment diagram of circuit;
Fig. 7 is the linear figure scheme drawing of track of the present invention image amount of information;
Fig. 8 is an obstacle grading forewarning system diagram of circuit of the present invention;
Fig. 9 shows the scheme drawing that the car body left side has obstacle to pass through for the present invention.
Description of reference numerals:
The 10-first image collection device; The 20-second image collection device; 30-the 3rd image collection device; The 40-processing unit; The 41-microprocessor; The 42-signal processor; 50-first annunciator; 60-second annunciator; The 100-Oversize vehicle; 110-speed of a motor vehicle sensor; The 120-brake system; The 130-turn signal system.
The specific embodiment
The present invention judges by the technology of image processing whether large-scale Chefei's direct vision district has hiding dangerous the existence, and sends the startup that different alarm signals are controlled image or light source or sound annunciator according to dangerous level.Therefore, before the explanation embodiment of the invention, earlier at being equipped with Oversize vehicle of the present invention to define its non-direct vision district scope.
In force, (Economic Commission for Europe ECE) stipulates N in rules the 46th clause with Economic Commission of Europe 3The Oversize vehicle of classification is an example.Its decree is pointed out, N 3The non-direct vision district in Oversize vehicle the place ahead of type is preceding 2 meters an of car, promptly belong to directly observation scope of steerman beyond 2 meters, again, in the non-direct vision zone of its car side also is 2 meters of car sides, and rise along the steerman position and to calculate 1 meter forward and the distance of calculating 1.75 meters backward, but if promptly belong to steerman direct viewing zone beyond exceeding this scope.After learning large-scale Chefei's direct vision zone, beneathly promptly make further detailed explanation at structure of the present invention and work flow.
At first consult Fig. 1 and Fig. 2, it is one embodiment of the invention scheme drawing, be equipped with Oversize vehicle 100 states of the present invention in order to expression, and for example this implementation system Organization Chart with Fig. 3 shows that native system includes one first image collection device 10, one second image collection device 20, one the 3rd image collection device 30, a processing unit 40, one first annunciator 50 and one second annunciator 60; And except said elements, present embodiment comprises a speed of a motor vehicle sensor 110, a brake system 120 and a turn signal system 130 that electrically connects Oversize vehicle 100.
Described image collection device 10,20,30 is generally charge coupled cell (CCD) or complementary metal oxide quartz conductor photographic goods such as (CMOS), it electrically connects processing unit 40, and the position that cooperates aforementioned Oversize vehicle 100 non-direct vision districts, the first image collection device 10 is erected at Oversize vehicle 100 headstock the place aheads, in order to produce one first non-direct vision district image, the second and the 3rd image collection device 20,30 then is erected at two sides about headstock respectively, in order to producing one second non-direct vision district image, and one the 3rd non-direct vision district image.Yet,, be not for definitely, when answering actual demand to do appropriateness adjustment about image collection Device Testing scope and quantity.
40 of processing units include a microprocessor 41 and a signal processor 42.Wherein, microprocessor 41 electrically connects speed of a motor vehicle perceptrons 110, brake system 120 and turn signal system 130, and after microprocessor 41 obtains the speed information that speed of a motor vehicle perceptron 110 transmits, produces a speed signal; Moreover microprocessor 41 also produces a brake signal when the brake system 120 of Oversize vehicle 100 starts, and when turn signal system 130 starts, produces a turn sign.Signal processor 42 in the present embodiment, be multichannel digital signal treater (multi-channel DSP), it can parallel processing multiple tracks signal source, set up the exploitation simulation environment, logic analysis, and the function of digital figure output, again, signal processor 42 electrically connects first, second, the 3rd image collection device 10,20,30 and microprocessor 41, and it is at the default one first pre-police region A of the image detection position in corresponding first non-direct vision district simulation, these first pre-police region A length and width are set at preceding 3 meters of car, and exceed vehicle body wide about each scope of 1 meter, simulation is provided with one second and one the 3rd pre-police region B in side region image detection position, the corresponding left and right sides in addition, C, this the second and the 3rd pre-police region B, the C length and width distinctly exceed 3 meters of car sides, and contain by pushing away 1 meter scope with 5 meters of pushers before the steerman position, use the aforementioned decree standard range of serious offense, form the effectiveness of caution in advance; Certainly the design of all data should be revised according to the difference of image collection Device Testing ability and car type, just can correspond to actual needs.
Moreover, after signal processor 42 receives all non-direct vision district images, to judge earlier that the obstacle whether the side travel amount is arranged in the first pre-police region of the place ahead occurs, whether there is the obstacle of the amount of vertically moving to occur in left and right sides second and the 3rd pre-police region, and confirming whether receive brake signal or turn sign after obstacle occurs, according to level hazard level weight, send one first alarm signal in regular turn or send one second alarm signal.And microprocessor 41 is the two-way communication platform with signal processor 42, the alarm signal that the vehicle speed signal that microprocessor 41 is received, brake signal, turn sign and signal processor 42 the are sent transmission of can communicating with each other between two platforms.
Mandatory declaration be, as shown in Figure 4, it also divides first annunciator 50 and makes first, second and third caution district a, b, c and match one to one with first, second and third pre-police region A, B, C work, use and near operator's saddle, utilize image to show that the zone appears in obstacle, and when first annunciator 50 is the LCD display screen, directly be wired to signal processor 42, be able to direct modeling caution district a, b, c situation output digitized video, if when it is luminous element such as LED lamp, then can electrically connect microprocessor 41; Again, second annunciator 60 is the element that buzzer phone or loudspeaker etc. have sound effect, and it electrically connects microprocessor 41.When first alarm signal starts, first annunciator 50 attracts the steerman warning with image or light source, when second alarm signal starts when with its startup, first annunciator 50 and second annunciator 60 will start simultaneously, and expression further forces steerman to note with sound.
After element of the present invention and element relation are being described, the particularly important is, how the present invention tells the obstacle that needs in non-direct vision district image, and the N/R background.The side after confirming obstacle, just can be the more further mode of clear interpretation this case flow process control.
Consult Fig. 5, suppose in an obstacle enters the first non-direct vision zone, and move to the G point by the F point, please cooperate with reference to shown in the barrier judgment diagram of circuit of figure 6, as shown in the figure, enter step S100 and start the collision caution system, also represent the startup of image collection device 10, in step S110, will import non-direct vision district image, then step S120 begins the image detection of non-direct vision district, image is in input this moment, and estimate the area image entropy (entropy) of the first non-direct vision district image by signal processor 42 at step S130, calculating at the track of its each time picture image amount of information, and store its result, the computing formula (1) of relevant track image amount of information is as follows:
I ( x ) = - Σ i = 1 n log ( p i ) × p i 30 - - - ( 1 ) ; And
p i = G i T pixels - - - ( 2 )
Wherein, the track image amount of information of the non-for this reason direct vision of I district image; G iFor gray value i number occurs in this non-direct vision district image; p iFor gray value i probability occurs in this non-direct vision district image; T PixelsValid pixel sum total in the non-for this reason direct vision district image; Again, find that through experiment when only having grey road surface and white lane mark to exist, the track image amount of information of this picture is on the low side, otherwise, in the picture that has pedestrian, bicycle or self-propelled vehicle, and will obvious crest in image, occur during the object side travel, and have higher image amount of information and occur, therefore, whether the linear figure of track image amount of information distributes as shown in Figure 7, behind the end step S130, can confirm to have obstacle to exist.So, input one predetermined threshold value Ts in signal processor 42 in advance, its effect is how alert signal treater 42 differentiates different between obstacle and the background, therefore after entering step S140, at first calculate the difference of the quantity of information of adjacent two pictures, compare with predetermined threshold value Ts again.Computing formula (3) about predetermined threshold value Ts is as follows:
T s = ( I max - I min ) 2 - - - ( 3 ) ;
Wherein, Ts is a predetermined threshold value, I MaxMaximum fault information when existing for obstacle is with I MinFor simple road image quantity of information, through this, if the quantity of information difference then is judged to be obstacle greater than threshold value Ts; After confirming through the obstacle of step S140, if the result is for being that expression promptly enters step S150 and notes down the obstacle position through confirming as obstacle; Otherwise, in step S140, obtain negative decision, promptly get back to step S110 and re-enter non-direct vision district, all the place aheads image.Moreover the step S140 also track image amount of information of two pictures before and after analysis time changes, also, obtain present time t and two pictures of preceding time t-1, and with this two pictures track image amount of information I tWith I T-1Subtract each other, can obtain the track image amount of information difference of adjacent two pictures by following formula (4);
ΔI t(x)=I t(x)-I t-1(x) (4);
Wherein, Δ I t(x) be track image amount of information difference; I t(x) with I T-1(x) then respectively be the track image amount of information of t time and t-1 time.
Again, when step S150, the obstacle position that signal processor 42 meetings were noted down according to each time, and according to behind the obstacle position reckoning motion track, the side travel amount of step S160 disturbance of analysis thing with and moving velocity, the formula of use (5) is just as following:
Behind end step S150, expression can estimate obstacle side travel amount and moving velocity; In addition, whether above stated specification has mobile object to occur in confirming picture, and this mobile object does not have and may be the judgement of obstacle, but, image detects and often is subjected to ambient environment factor affecting judged result, even more serious at open space, so have the problem generation that too sensitive sensing result is grown the steerman puzzlement on the contrary unavoidably, therefore, this example assumes automobile, locomotive, side travel speed in obstacle such as bicycle or the pedestrian short time all should be in 22.5km/h, therefore to decide 22.5km/h be the obstacle one moving velocity upper limit in the present invention, establish a predetermined speed in also promptly in processing unit, this predetermined speed is 22.5km/h, if after the calculating of formula (5), the side travel speed of obstacle exceeds this standard, among the step S160, will judge that this mobile object is not to be real obstacle, just enter step S180 and do not do any follow-up caution action; Otherwise, if step S160 confirms that the side travel speed of obstacle is lower than this and moves the upper limit again, just in 0~22.5km/h, represent that it should be the obstacle that desire is followed the trail of, then entering among the step S170 for this reason, obstacle carries out the early warning operation.In addition, no matter be step S170 or step S180, after it finishes, all should get back to step S110 repetition institute and carry out in steps.
Through step S100 to the analysis of step S180, method that should cheer and bright barrier judgment of the present invention; Again, the state of general Oversize vehicle 100 the most normal generation harm, be also not discover headstock the place ahead when just belonging to the low speed starting to pass through object, and limited car body hovel influence when changeing about vehicle body, do not discover two kinds of car body left and right sides mobile objects yet, therefore the present invention is directed to this two reason, the method for inspection of signal processor 42 is divided into the condition that non-direct vision district, the place ahead is detected, and the condition that the non-direct vision of left and right sides district is detected illustrates one by one.
Consult Fig. 8, be obstacle grading forewarning system diagram of circuit of the present invention.At first, shown in step S200, after the collision caution system starts, image collection device 10,20,30 will be opened; When step S210, signal processor 42 receives the place ahead and the left and right non-direct vision district image of this three image collections device 10,20,30 simultaneously; Then, enter step S220, have the road speed restriction of Oversize vehicle 100 in the signal processor 42, in originally being embodied as 20km/h, differentiate Oversize vehicle 100 present driving states by this, and when road speed is lower than 20km/h, select to enter step S230, and when road speed is higher than 20km/h, select to enter step S240; Suppose that step S220 judges that the road speed of Oversize vehicle 100 is lower than 20km/h, when step S230, import aforesaid detection of obstacles flow process (S100-S180), use and confirm whether to have among the pre-police region A in the place ahead obstacle to exist, when the place ahead that is judged as of step S230 has under the situation of obstacle, confirm whether to have the brake signal of receiving that microprocessor 41 is sent here, when the result also is "Yes", select step S231, represent the place ahead to have object to pass through, but this Oversize vehicle 100 is braking, and the speed of a motor vehicle is lower than 20km/h, therefore as long as send first alarm signal at step S231, just utilize the first pre-police region a of display screen 50 to send light source, inform that the steerman obstacle still must add to note the place ahead situation just forwardly; Otherwise, when the place ahead that is judged as of step S230 has under the situation of obstacle, but there is not brake signal from microprocessor 41, selection enters step S231, expression the place ahead has object to pass through, though this car speed of a motor vehicle is lower than 20km/h, but because the driver does not touch on the brake, the danger so bump easily, so send second alarm signal at step S234,, more utilize second annunciator 60 to sound and attract steerman to note just except the first pre-police region a display light source of display screen 50, remind steerman to dodge or to touch on the brake, avoid dangerous and take place.Again, step S232 and step S234 enter among the step S240 after finishing without exception.
Step S230~S234 has illustrated the collision avoidance system flow process in headstock the place ahead, be noted that, reason based on Oversize vehicle 100 car weights and vehicle commander, and the place ahead of the present invention pre-police region A scope only reaches and exceeds the preceding 3 meters distance of car, in other words, and when speed is higher than 20km/h, if really detecting obstacle appears in the pre-police region A in the place ahead, also have little time brake, if brake reluctantly is afraid of to take place more serious accident again.Come, the present invention carries out the flow process that the early warning of the car body left and right sides detects when explanation again.Get back to the step S220 of Fig. 8, when judging that at step S220 road speed surpasses 20km/h, the detection of obstacles of two sides about the beginning car body; At first, whether judge earlier when step S240 has the obstacle of the amount of vertically moving to occur in the second pre-police region B of left side, if as shown in Figure 9, there is vehicle to appear among the pre-police region B in left side, at this moment, confirm to have or not at present to receive turn sign when entering step S241 again, suppose to receive turn sign, when judging whether wherein to carry left-hand rotation information secretly, that is to say, if steerman does not plan to turn to, then as long as send left side first alarm signal at step S242,50 second caution district b send color light source at display screen, the notice steerman is noted, otherwise, when confirm at step S241 steerman will left-handed turning to the time because the side swipe chance significantly increases, therefore should send left side second alarm signal at step S243, with aforementioned all identical, promptly be to send color light source at the display screen second caution district b, also send the sound simultaneously and remind steerman to note by second annunciator 60.Again, occur in step S240 hypothesis and clear, in entering step S250, the detection of obstacles of beginning right side the 3rd pre-police region C, then confirm whether to receive turn sign,, then send a right side first alarm signal at step S252 if do not have at step S251, if have, when send right side second alarm signal at step S253; Because step S250-S253 is identical with determination methods and the principle of step S240~S243, so do not giving unnecessary details.
By aforesaid vehicle collision alarming system, can initiatively remind the driver, crash-avoidance is unexpected to be taken place.Detect and under image processing technique complements each other by image, as long as have mobile object in the non-direct vision district image, all can confirm to make and it will be set at the obstacle that to follow the trail of via data processing technique, moreover, the present invention is more in conjunction with the speed sensor of Oversize vehicle, system is considering that the Oversize vehicle quality is heavy, the factor that braking distance is long, when carrying out the subregion sensing according to speed signal, in addition, the present invention is again in conjunction with the brake system and the turn signal system of Oversize vehicle, for doing the hazard level classification in the position of obstacle, initiatively under critical situation, utilize sound and light program to cause that synchronously steerman notes, and under general situation, then take care in the zone of reminding steerman to be careful with picture or light source.
The above description of this invention is illustrative, and nonrestrictive, and those skilled in the art is understood, and can carry out many modifications, variation or equivalence to it within spirit that claim limits and scope, but they will fall within the scope of protection of the present invention all.

Claims (10)

1. vehicle collision alarming system at least one non-direct vision district of an Oversize vehicle carrying out detection of obstacles, is characterized in that this vehicle collision alarming system comprises:
At least one image collection device, it installs on this Oversize vehicle, and this image collection device is gathered at least one this non-direct vision district image,
One processing unit, it electrically connects this image collection device to differentiate at least one obstacle, have a predetermined speed in this processing unit, change according to this obstacle position is relative with the time, calculate the moving velocity and the amount of movement of this obstacle, this processing unit is preset with at least one pre-police region that is arranged in this non-direct vision district image, when this obstacle of assessment has amount of movement and this moving velocity when being lower than this predetermined speed in this pre-police region, sends at least one early warning signal; And
At least one annunciator, it electrically connects this processing unit, and this annunciator receives this early warning signal and produces caution sound or light source or image.
2. vehicle collision alarming system as claimed in claim 1, it is characterized in that, this processing unit set this predetermined speed 0~22.5 kilometer/time between, and calculate the track image amount of information of this non-direct vision district each picture of image, and the difference of calculating the track image amount of information of continuous two these pictures, this processing unit is got the variation of this difference curve and the predetermined threshold value in this processing unit compares, if this difference has and is higher than this predetermined threshold value, and the moving velocity position is between this predetermined speed range, when this difference being exceeded set positions for this obstacle, and this difference is lower than the position of this predetermined threshold value and set positions that moving velocity exceeds this predetermined speed range is a background.
3. vehicle collision alarming system as claimed in claim 2 is characterized in that this processing unit more includes a signal processor, and in order to estimate the track image amount of information of this non-direct vision district image, its formula is calculated as follows:
I ( x ) = - Σ i = 1 n log ( p i ) × p i ;
p i = G i T pixels ; And
Wherein, I is arbitrary this track image amount of information; G iFor number appears in gray value i in area image; p iFor probability appears in gray value i in area image; T PixelsBe valid pixel sum total in the area image.
4. vehicle collision alarming system as claimed in claim 3 is characterized in that, this signal processor calculates this predetermined threshold value with following formula:
T s = ( I max - I min ) 2 ;
Wherein, T sBe predetermined threshold value, I ManMaximum fault information when existing for obstacle is with I MinBe simple road image quantity of information.
5. vehicle collision alarming system as claimed in claim 3 is characterized in that, this signal processor calculates this obstacle this amount of movement in continuous this two picture with following formula:
ΔI t(x)=I t(x)-I t-1(x);
Wherein, Δ I t(x) be this obstacle in continuous time, this amount of movement of this two picture; I t(x) with I T-1(x) then respectively be the track image amount of information of t time and t-1 time.
6. vehicle collision alarming system as claimed in claim 5 is characterized in that, this microprocessor utilizes the ratio of miles of relative movement and actual distance in the pixel, and the variation of relative time, calculates true amount of movement of this obstacle and moving velocity:
7. vehicle collision alarming system as claimed in claim 1 is characterized in that, this Oversize vehicle the place ahead and left and right sides respectively have at least one this non-direct vision district; This at least one image collection device includes one first image collection device, and it is installed on this Oversize vehicle the place ahead, in order to one first non-direct vision district image is obtained in this non-direct vision zone in the place ahead; One second image collection device, it is installed on this Oversize vehicle left side, in order to one second non-direct vision district image is obtained in this non-direct vision zone in left side; And one the 3rd image collection device, it is installed on the right side of this Oversize vehicle, in order to one the 3rd non-direct vision district image is obtained in this non-direct vision zone on right side; And this non-direct vision position is when this Oversize vehicle the place ahead, and this amount of movement is a side travel amount; When this non-direct vision position during on this full size vehicle left side or right side, this amount of movement is the amount of vertically moving.
8. vehicle collision alarming system as claimed in claim 7 is characterized in that, at least one pre-police region of this of this processing unit comprises one first pre-police region, and its position is in this first non-direct vision district image; One second pre-police region, it is positioned at this second non-direct vision district image; And one the 3rd pre-police region, it is in the 3rd non-direct vision district image; This annunciator has one first caution district to should the first pre-police region, and one second caution district is to should the second pre-police region, and one the 3rd warn and distinguish should the 3rd pre-police region, shows this barrier region in order to light source or image.
9. vehicle collision alarming system as claimed in claim 8 is characterized in that, this Oversize vehicle has a speed of a motor vehicle perceptron, a brake system and a turn signal system; And this processing unit comprises that more a microprocessor electrically connects this speed of a motor vehicle perceptron, this brake system and this turn signal system, this microprocessor sees through this speed of a motor vehicle perceptron and obtains a speed signal, see through this brake system and obtain a brake signal, and obtain a turn sign through this turn signal system; Be provided with road speed restriction in this microprocessor, this road speed be restricted to 20 kilometers/time, when this speed signal limits greater than this road speed, this microprocessor is abandoned this first pre-police region and is detected, only detect this second pre-police region and the 3rd pre-police region, and cooperate this turn sign to make the grading forewarning system notice at this second pre-police region or the 3rd pre-police region; When this speed signal during less than the restriction of this road speed, carry out detection of obstacles at this first pre-police region earlier, and cooperate this brake signal to make grading forewarning system to notify, follow-uply detect at this second pre-police region and the 3rd pre-police region again.
10. as vehicle collision alarming system as described in the claim 9, it is characterized in that this at least one annunciator comprises one first annunciator and one second annunciator; This microprocessor can send one first early warning signal and start this first annunciator effect, and sends one second early warning signal and start this first annunciator and the second annunciator effect; And this image collection device is charge coupled cell or complementary metal oxide quartz conductor video camera, and this annunciator can be buzzer phone or LED lamp or LCD screen and voice system.
CN200910142283A 2009-06-29 2009-06-29 Vehicle collision warning system Active CN101934771B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN200910142283A CN101934771B (en) 2009-06-29 2009-06-29 Vehicle collision warning system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN200910142283A CN101934771B (en) 2009-06-29 2009-06-29 Vehicle collision warning system

Publications (2)

Publication Number Publication Date
CN101934771A true CN101934771A (en) 2011-01-05
CN101934771B CN101934771B (en) 2012-10-03

Family

ID=43388416

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200910142283A Active CN101934771B (en) 2009-06-29 2009-06-29 Vehicle collision warning system

Country Status (1)

Country Link
CN (1) CN101934771B (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102303606A (en) * 2011-06-17 2012-01-04 清华大学 Hazard assessment method based on retarded velocity required by vehicle
TWI426470B (en) * 2011-09-20 2014-02-11 Univ Chaoyang Technology Corner collision warning device
CN103625405A (en) * 2012-08-24 2014-03-12 能晶科技股份有限公司 Driving assisting system and starting method thereof
CN103650015A (en) * 2011-07-21 2014-03-19 丰田自动车株式会社 Vehicle information transmitting apparatus
CN104166858A (en) * 2014-08-06 2014-11-26 张锦奇 Anti-collision method and system for high-speed movement of objects
CN104299439A (en) * 2013-07-15 2015-01-21 大众汽车有限公司 Device and method for displaying traffic situation in vehicle
CN104512325A (en) * 2013-09-12 2015-04-15 罗伯特·博世有限公司 Object differentiation warning system
CN104718565A (en) * 2012-11-27 2015-06-17 日产自动车株式会社 Vehicle acceleration restriction device and vehicle acceleration restriction method
CN105554346A (en) * 2015-12-14 2016-05-04 宁波裕兰信息科技有限公司 Mobile object detection system based on vehicle-mounted video
CN105575183A (en) * 2014-01-26 2016-05-11 浙江吉利控股集团有限公司 Vehicle travel early-warning method
CN106935077A (en) * 2017-03-17 2017-07-07 合肥工业大学 One safety pre-warning system with auxiliary of parking is detected based on vehicle blind spot
CN106997049A (en) * 2017-03-14 2017-08-01 奇瑞汽车股份有限公司 A kind of method and apparatus of the detection barrier based on laser point cloud data
CN110274542A (en) * 2018-03-15 2019-09-24 艾沙技术有限公司 Mobile carrier, safety alarm device and safety alarm method
TWI688502B (en) * 2018-02-14 2020-03-21 先進光電科技股份有限公司 Apparatus for warning of vehicle obstructions

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4186908B2 (en) * 2004-10-28 2008-11-26 アイシン精機株式会社 Moving object periphery monitoring device
JP4720386B2 (en) * 2005-09-07 2011-07-13 株式会社日立製作所 Driving assistance device
TWI269727B (en) * 2006-01-09 2007-01-01 Ind Tech Res Inst Method and apparatus of assistant monitor for vehicle
JP5347257B2 (en) * 2007-09-26 2013-11-20 日産自動車株式会社 Vehicle periphery monitoring device and video display method

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102303606A (en) * 2011-06-17 2012-01-04 清华大学 Hazard assessment method based on retarded velocity required by vehicle
CN103650015A (en) * 2011-07-21 2014-03-19 丰田自动车株式会社 Vehicle information transmitting apparatus
US9384649B2 (en) 2011-07-21 2016-07-05 Toyota Jidosha Kabushiki Kaisha Vehicle information transmitting apparatus
TWI426470B (en) * 2011-09-20 2014-02-11 Univ Chaoyang Technology Corner collision warning device
CN103625405A (en) * 2012-08-24 2014-03-12 能晶科技股份有限公司 Driving assisting system and starting method thereof
CN103625405B (en) * 2012-08-24 2015-12-16 能晶科技股份有限公司 Driving ancillary system and enable method
CN104718565A (en) * 2012-11-27 2015-06-17 日产自动车株式会社 Vehicle acceleration restriction device and vehicle acceleration restriction method
CN104299439A (en) * 2013-07-15 2015-01-21 大众汽车有限公司 Device and method for displaying traffic situation in vehicle
CN104512325B (en) * 2013-09-12 2018-10-26 罗伯特·博世有限公司 Object distinguishes caution system
CN104512325A (en) * 2013-09-12 2015-04-15 罗伯特·博世有限公司 Object differentiation warning system
CN105575183A (en) * 2014-01-26 2016-05-11 浙江吉利控股集团有限公司 Vehicle travel early-warning method
CN104166858B (en) * 2014-08-06 2018-03-23 张锦奇 A kind of collision-proof method and system of objects at high speed motion
CN104166858A (en) * 2014-08-06 2014-11-26 张锦奇 Anti-collision method and system for high-speed movement of objects
CN105554346A (en) * 2015-12-14 2016-05-04 宁波裕兰信息科技有限公司 Mobile object detection system based on vehicle-mounted video
CN106997049B (en) * 2017-03-14 2020-07-03 奇瑞汽车股份有限公司 Method and device for detecting barrier based on laser point cloud data
CN106997049A (en) * 2017-03-14 2017-08-01 奇瑞汽车股份有限公司 A kind of method and apparatus of the detection barrier based on laser point cloud data
CN106935077A (en) * 2017-03-17 2017-07-07 合肥工业大学 One safety pre-warning system with auxiliary of parking is detected based on vehicle blind spot
TWI688502B (en) * 2018-02-14 2020-03-21 先進光電科技股份有限公司 Apparatus for warning of vehicle obstructions
US10812782B2 (en) 2018-02-14 2020-10-20 Ability Opto-Electronics Technology Co., Ltd. Obstacle warning apparatus for vehicle
CN110274542A (en) * 2018-03-15 2019-09-24 艾沙技术有限公司 Mobile carrier, safety alarm device and safety alarm method

Also Published As

Publication number Publication date
CN101934771B (en) 2012-10-03

Similar Documents

Publication Publication Date Title
CN101934771B (en) Vehicle collision warning system
CN106240458B (en) A kind of vehicular frontal impact method for early warning based on vehicle-mounted binocular camera
US10332401B2 (en) Running vehicle alerting system and method
CN100440269C (en) Intelligent detecting prewarning method for expressway automobile running and prewaring system thereof
CN102361777B (en) Device for generating sound to outside of vehicle
US8310355B2 (en) Vehicular collision warning system
CN104217590A (en) On-board traffic density estimator
US9524643B2 (en) Orientation sensitive traffic collision warning system
JP4097519B2 (en) Danger sensitivity estimation device, safe driving evaluation device and alarm device
CN106530831A (en) System and method for monitoring and early warning of high-threat vehicles
CN106573589B (en) Collision prediction time computing device and collision prediction time calculation method
CN104085396A (en) Panoramic lane departure warning method and system
CN107672551B (en) Front vehicle anti-collision early warning system and control method thereof
AU2007201110A1 (en) Systems and methods for determining threshold warning distances for collision avoidance
CN101391589A (en) Vehicle intelligent alarming method and device
CN106611513B (en) Traffic information prediction device, system and method
CN105015411B (en) A kind of automobile microwave radar anti-collision early warning method and system based on video fusion
US10008118B2 (en) Vehicle collision avoidance system and method
CN106097776A (en) A kind of bend collision early warning system and control method thereof
CN104299400A (en) Vehicle environment monitoring system
US8184018B2 (en) Image-based vehicle safety warning system
CN109949612B (en) Ghost probe accident early warning system based on information interaction
JP5362225B2 (en) Operation recording device and operation status recording method
CN106183986A (en) A kind of intelligent driving safety system and method
CN109895694B (en) Lane departure early warning method and device and vehicle

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model