EP0877944A1 - Optisches entfernungs-und geschwindigkeitsmessystem - Google Patents
Optisches entfernungs-und geschwindigkeitsmessystemInfo
- Publication number
- EP0877944A1 EP0877944A1 EP95910109A EP95910109A EP0877944A1 EP 0877944 A1 EP0877944 A1 EP 0877944A1 EP 95910109 A EP95910109 A EP 95910109A EP 95910109 A EP95910109 A EP 95910109A EP 0877944 A1 EP0877944 A1 EP 0877944A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- target
- line
- camera
- speed
- offset
- 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.)
- Withdrawn
Links
- 230000003287 optical effect Effects 0.000 title claims abstract description 73
- 238000001514 detection method Methods 0.000 title description 6
- 238000003491 array Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 54
- 238000004364 calculation method Methods 0.000 claims description 27
- 238000005259 measurement Methods 0.000 claims description 18
- 238000003708 edge detection Methods 0.000 claims description 13
- 238000012417 linear regression Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims 2
- 101150034459 Parpbp gene Proteins 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 26
- 230000008569 process Effects 0.000 description 11
- 230000006870 function Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 5
- 238000012546 transfer Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 4
- 230000033001 locomotion Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- XUKUURHRXDUEBC-KAYWLYCHSA-N Atorvastatin Chemical compound C=1C=CC=CC=1C1=C(C=2C=CC(F)=CC=2)N(CC[C@@H](O)C[C@@H](O)CC(O)=O)C(C(C)C)=C1C(=O)NC1=CC=CC=C1 XUKUURHRXDUEBC-KAYWLYCHSA-N 0.000 description 1
- JEYCTXHKTXCGPB-UHFFFAOYSA-N Methaqualone Chemical compound CC1=CC=CC=C1N1C(=O)C2=CC=CC=C2N=C1C JEYCTXHKTXCGPB-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C3/00—Measuring distances in line of sight; Optical rangefinders
- G01C3/02—Details
- G01C3/06—Use of electric means to obtain final indication
- G01C3/08—Use of electric radiation detectors
- G01C3/085—Use of electric radiation detectors with electronic parallax measurement
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P3/00—Measuring linear or angular speed; Measuring differences of linear or angular speeds
- G01P3/36—Devices characterised by the use of optical means, e.g. using infrared, visible, or ultraviolet light
Definitions
- the following invention relates to a system for ranging and speed detection, and more particularly relates to an optical system for determining range and speed that is passive and does not require an energy transmitter to accomplish its purpose.
- Radar is a primary example of this technique, and radar guns are conven ⁇ tionally used by law enforcement agencies for traffic control.
- the problem with radar as a traffic control device is that target acquisition and measurement are ambiguous. It can frequently not be determined which target out of a multitude of possible targets is responsible for generating any particular speed indica ⁇ tion.
- Another problem is that radar can be detected by receivers tuned to the proper frequency. Laser ranging systems are also available but such systems are also detectable at the target and are prohibitively expensive.
- a passive optical speed and distance measuring system which includes a pair of video camera lenses pointed along respective lines of sight towards a target.
- the lenses are positioned along a common baseline a prede ⁇ termined distance apart.
- a timer causes both camera lenses to capture a first target image in the field of view of each lens at a first time T, and also at a later time T 2 .
- the images are transferred to a range measuring means which determines a distance R. from the common baseline to the target at a time T 1 and for determining the distance R 2 from the baseline to the target at a time T 2 .
- the lines of sight of each of the camera lenses are parallel, and the lenses may be included in separate cameras, or, through the use of mirrors and prisms, each lens may be focused on a common image detec ⁇ tion device such as a charge coupled device (CCD) .
- the cameras are generally of the type that provide an image which is electronically scanned in a pixel array of light sensitive devices where the light intensity values of the pixels in the array may be digitized and stored for processing by a computer.
- the first and second camera lenses each focus images found in their respective fields of view on a pixel map. Contained in the pixel map is at least a partial image of the target which is found in the field of view of each camera lens.
- the range measuring means includes comparison means for correlating the position of the target image on the .first pixel map with the same corresponding target image on the pixel map of the second camera lens. This correlation is used to determine the angle between the target and the respective lines of sight of the first and/or second camera lenses.
- the pixel map may comprise a single line scan of video data and the comparison means includes means for determining the differences of video data intensities at each respective pixel location in the respective pixel maps to find the correlation between the target images. This correlation will occur at a global null which will be the position giving a minimum intensity difference between the pixel maps. This pixel location is a linear function of the angle between the target image and the line of sight of at least one of the cameras.
- the video data contained in the pixel map may be converted to digital data for processing by a computer program.
- the computer program calculates the absolute value differences in intensity between pixels in the pixel maps from the first and second video cameras at successively offset shift positions between the two pixel maps.
- preprocessing circuitry may be used to shift the pixel maps relative to each other to create a list of the values at all possible pixel offsets. The computer may then search the list for the global null.
- the first and second lenses may be mounted in a camera array which further includes a third camera for sighting and target acquisition.
- the first and second cameras generally should have narrow fields of view to avoid optical background clutter, but the third camera may have a wide field of view for target acquisition. Additionally, all three cameras may be mechanically connected so that the first and second camera lenses may be slaved to rotational motion by the third (target acquisition) camera.
- the third camera may further include an alphanumeric video generator for providing predetermined message data upon acquisition of a selected target by an operator. Such data could include, for example, the time and date and other information about the target.
- Alternative methods of determining the offset angle of the target image relative to the line of sight may include detecting the edges of the target as the target changes its position relative to the optical back ⁇ ground. Since the background does not change over most of the field of view as the target moves, signal process ⁇ ing means may be used to discriminate between a moving edge and static background video data.
- the cameras and/or camera lenses are mounted side-by-side along a horizontal line which is parallel to the line scan of the light sensitive device in the focal plane of the lenses.
- the light sensitive element may be oriented so that the line scan direction is perpendicular to the baseline.
- the signal processing to determine the target image offset from the line of sight in this case is slightly different from that which is used when the light sensi- tive element is scanned parallel to the baseline.
- a video line scan of one of the cameras serves as a "template” and whole line scans of the other camera are compared to the template line to establish image correlation, thus determining the offset. This method is more accurate than determining the offset by comparison of overlapped fractions of a line but requires more memory and computational time.
- Determination of the difference in pointing angles to the target between the two cameras allows a computer associated with this system to accurately estimate the distance to the target.
- the computer does this by solving the essential trigonometry of the camera configuration. Because the two camera lenses are mounted along a common baseline a predeter ⁇ mined distance apart, lines extending from each of the camera lenses to the target form a triangle. With both offset angles known, a trigonometric solution that solves for the distance from the target to the center of the baseline can be calculated. This measurement is per ⁇ formed at least two different times and the difference in range to the target as a function of the time difference may be used to calculate the speed of the target.
- FIG. 1 is a simplified block schematic diagram of the system of the invention.
- FIG. 2 is a block schematic diagram of the video camera subsystem shown in FIG. 1.
- FIG. 3 is a block schematic diagram of the control and computational subsystem shown in FIG. 1.
- FIG. 4 is a simplified flow chart diagram of a computer program used to calculate the range and speed of a moving target.
- FIG. 4a is a schematic illustration of the optical relationships in the system using separate cameras with the line scans parallel to the baseline.
- FIG. 5 is a schematic illustration of the geometry of pixel maps which are used to calculate the angles between the target center and the lines of sight of the camera lenses.
- FIG. 6 is a schematic diagram of the overall geometry of the optical range and speed detecting system of the present invention.
- FIG. 7 is a schematic diagram illustrating the relationship between the system geometry of FIG. 6 and the pixel maps of FIG. 5.
- FIG. 8 is a schematic diagram illustrating the method by which the system calculates the target offset.
- FIG. 9a-9f are a flow chart diagram illustrating the method of calculating the range and speed to the target using the system of FIG. 4.
- FIG. 10 is a waveform diagram illustrating a second method of determining an offset angle between the cameras and the line of sight, termed the edge detection method.
- FIGS, lla-llc are a flow chart diagram illustrating the edge detection capture method illustrated in FIG. 10.
- FIG. 12 is a schematic view of the fields of view of right and left camera lenses arranged in the system illustrated in FIG. 7.
- FIG. 13 is a schematic diagram of overlapping fields of view of camera lenses arranged so that the CCD line scan is perpendicular to a vertical baseline.
- FIG. 14 is a schematic representation of the pixel frame maps produced by camera lenses arranged along a vertical baseline as illustrated in FIG. 13.
- FIG. 15 is a schematic representation of the geometry of the perpendicular line scan camera arrangement with upper and lower camera lenses oriented along a vertical baseline.
- FIG. 16 is a schematic flow chart diagram illustrating a method of calculating the pixel offset when the cameras are arranged along the vertical baseline as illustrated in FIG. 15.
- FIG. 17 is a schematic diagram in perspective view showing an embodiment of an invention using a single camera having dual side-by-side lenses scanning the CCD parallel to the baseline.
- FIG. 18 is a block schematic diagram of a video preprocessor to be used in an alternative embodiment of the invention.
- FIG. 19 is a schematic diagram showing a perspective view of another embodiment of the invention using a single camera having dual lenses mounted for scanning the CCD perpendicular to the baseline.
- FIG. 20 is a flow chart diagram similar to FIG. 4 illustrating a linear regression method for calculating speed.
- FIG. 21 is a schematic diagram of a two-dimensional memory array containing range and time data.
- FIG. 22 is a flow chart diagram illustrating a linear regression calculation for the method shown in FIG. 20.
- the invention includes a video camera subsystem and video display 10 connected to a control and computational subsystem 12.
- the camera subsystem 10 provides left- and right-hand camera video to the control subsystem 12 and the control subsystem supplies alphanumeric video to the video camera sub- system.
- FIG. 2 which shows an expanded block diagram of the video camera subsystem includes a narrow field-of- view lens 14 which provides an optical image to a left- hand video camera 16.
- a second narrow field-of-view camera lens 18 provides an optical image to a right-hand master video camera 20.
- the right-hand video camera includes sync which is supplied to the left-hand video camera 16 and to a sighting video camera 22 which is also slaved to the right-hand video camera 20.
- the sighting video camera 22 includes a wide field-of-view lens 24 for target acquisition.
- All of the cameras 16, 20 and 22 are of the type that include a pixel matrix array of light sensitive devices such as a CCD. As such the pixel array is scanned electronically in horizontal or vertical lines to provide video data which represents light intensity at each scanned pixel location.
- the output of the sighting video camera 22 provides video to a video mixer 26 which may receive an alphanumeric video input.
- the wide field-of-view image of the camera 22 along with the alphanumeric video may be displayed on a video display 28.
- FIG. 3 is an expanded view of the control and computational subsystem 12.
- the subsystem 12 consists primarily of a personal computer 30 (shown in dashed outline) which may be any type of IBM ⁇ -compatible per ⁇ sonal computer.
- the computer 30 includes frame grabbers 32 and 34 for the left-hand camera video and the right- hand camera video, respectively.
- An alphanumeric gener ⁇ ator 36 is slaved to the sync provided by the right-hand video camera 20.
- the computer includes a computer bus 38 which couples the frame grabber 32, the alphanumerics generator 36 and the frame grabber 34 to a disk memory 40, random access memory 42 and a CPU/10 unit 44. External to the computer 30 the computer bus 38 is also coupled to a real time clock and calendar 46, operator's control unit 48 and a printer 50.
- the system geometry is illustrated in FIG. 6.
- the left-hand camera lens 14 is situated at a point 52 and the right-hand camera lens 18 is situated at a point 54.
- the wide field-of-view lens 24 is located midway between points 52 and 54 at a point 56.
- the target is located at a point 58 in the field of view of all three cameras.
- Preferably the respective lines of sight of the narrow field-of-view cameras located at points 52 and 54 are parallel as indicated by the dashed lines in FIG. 6.
- An angle ⁇ lH indicates the angle between the target point 58 and the line of sight of the camera located at 52.
- ⁇ RH indicates the angle between the line of sight and the target for the right-hand camera.
- the two cameras are situated along a baseline 60 having a length "b" where the center wide field-of-view camera lens 14 is located midway between the two narrow field-of-view cameras 14, 18 at a distance b/2 along the baseline. From this point the range to the target point 58 is indicated as "R.”
- a triangle having sides “a,” “b” and “c” is thus formed between the target location 58 and the right- and left-hand camera locations 52 and 54. This triangle includes internal angles ⁇ (alpha) at the left- hand camera location, ⁇ (gamma) at the right-hand loca ⁇ tion and ⁇ (beta) at the target location.
- the system determines the range R to the target by solving the trigonometry of the aforementioned triangle. In order to determine the speed of the target, it performs this calculation at two different times and divides the difference in range by the elapsed time between measurements.
- FIG. 4 illustrates in flow chart diagram form the essential system operational method.
- the system After “START,” the system, at block 62, captures a linear stereoscopic image of the target. This image consists of a single line scan of video data captured by the left-hand and right-hand lenses simultaneously. This linear scan is conducted at times T, and T 2 .
- the system calculates the sums of the differences in pixel intensity between the right-hand and left-hand scan lines, respec ⁇ tively, at each of the two times for the total possible number of pixel offset positions between the two scan lines.
- the system performs a search of the sums of differences list for nulls and a null list i ⁇ made.
- the system searches for a global null and records the pixel offset location at which it was found.
- a global null is the offset location that yields the minimum intensity difference between images. This offset location is proportional to an angle between the line of sight of at least one camera and the target.
- the system calculates the range to the target based upon the number of pixels offset required to achieve the global null.
- the process is repeated for the second line pair scan and at block 74 the system calculates the target speed by solving the equation:
- FIG. 4a illustrates the field of view overlap region at the range to the target, R (defined by points S, P, E and M) .
- WR is the width of this region.
- the width, WR is a fraction of a line, approaching unity as the range approaches infinity.
- M is the absolute minimum operating range (Rmin) .
- Rmin the absolute minimum operating range
- FIG. 4a where the upper pixel scan 76 repre ⁇ sents the left-hand camera lens pixel map and the lower scan 78 represents the right-hand camera lens pixel map.
- the dark pixel in each scan represents the calculated location of the target and the shift of distance 2d is the offset position where a global null occurs.
- This embodiment of the invention presumes that at least a portion of the target can be found within the field of view common to both lenses. This is shown graphically in FIG. 4A.
- FIG. 7 represents a special case for the generic system geometry shown in FIG. 6. Referring again to FIG. 5, the system determines the distance d which is one-half the pixel offset between the pixel maps 76 and 78. It does this by correlating the two lines.
- Correla ⁇ tion occurs at a total offset 2d as shown in the lower portion of FIG. 5 where the right-hand camera lens pixel map 78a has been offset with respect to the left-hand camera lens pixel map 76a to produce a total offset equal to 2d.
- pixel intensities are compared at this off ⁇ set, a global null will be produced which represents a minimum in the absolute differences in pixel intensities between the pixel maps at this offset location.
- the reason for this is that at the global null the target image is the dominant object common to both camera fields of view.
- the process is akin to shifting one of two overlapping fields of view visually to produce a coinci ⁇ dent field of view.
- the pixel offset needed to achieve this is then directly proportional to the angle between at least one of the camera lenses and the target image relative to its original line of sight.
- FIG. 7 A ray from the target 58a at point P makes an angle ⁇ between line c and the left-hand camera lens line of sight. This ray passes through the left-hand camera lens 80 and is focused on a pixel in the line pixel map 76 located in the focal plane 82 of the lens 80. Simi ⁇ larly, on the right side, a ray from point P on the target 58a extends along line "a" to a line pixel map 84 which is located in the focal plane 86 of the right-hand camera lens 88. Both pixel maps are located in their respective focal planes at the focal length "FL" of lenses 80 and 88 which are identical.
- the ray from the target extending along line "c" or line "a” makes an angle ⁇ with the camera line of sight.
- This angle is the same angle as the angle between the lines c and a, respectively, and line R which represents the range to the target 58A.
- the calculation step outlined in block 70 of FIG. 4 calculates the range to the target by first determining the value of angle ⁇ .
- Angle ⁇ is determined by finding the shift, in number of pixels, at which the absolute differences in pixel intensities between respective pixels and pixel map are at a minimum. This is equivalent to rotating the right- hand pixel map about an axis at the right-hand lens 88 corresponding to an angle ⁇ which is equal to two times ⁇ .
- This imaginary rotation of the pixel map is shown graphically in FIG. 7 wherein pixel map 8 A has been rotated through an angle ⁇ so that the dark (target image) pixel in each map is at the same location.
- the dashed line at lens 88A indicates the revised "rotated" position of the right-hand camera lens.
- FIGS. 9a through 9f a flow chart diagram is shown which describes the method by which the control and computational subsystem measures the target speed.
- computer software under the control of the CPU/10 unit 44 in the computer 30 enters a loop (blocks 87, 89) which waits for a "measure speed" command.
- the command is initiated by the operator's control unit 48 which may include a simple push button switch associated with the wide angle field- of-view camera 22.
- a "command received" flag is set (block 90) which initiates the simultaneous capture of video in the left-hand and right- hand frame grabbers 32 and 34, respectively (block 92).
- the video lines that are captured are transferred from each frame grabber 32, 34 to the RAM 42 in the computer 30 (block 94) .
- both video lines are tagged with a time T 1 representing a time of capture (block 96) .
- the program loops back to block 92 to repeat the sequence (blocks 98 and 100) .
- two video line-pairs are stored in computer RAM tagged with times T, and T 2 , respectively.
- the software sets a variable OS (offset) equal to zero (block 102) .
- Another variable, ACC is also set equal to zero (block 104) .
- a third variable, PIX is made equal to OS plus 1 (block 106) .
- the accumulator variable, ACC is then made equal to ACC plus an expression which represents the absolute value of the difference in intensities between the left-hand pixel map and the right-hand pixel map at a predetermined offset position (block 108) .
- the PIX variable is then incre ⁇ mented by one (block 110) and a loop is formed so that the calculation will be repeated until the variable PIX is equal to NPIX which represents the total number of pixels in the line of video (block 112) . In a conven ⁇ tional video camera of this type there will be 511 pixels in a single horizontal line of video.
- QU01 (OS,l) which is made equal to the offset OS (block 114) .
- a second new variable QUOl (OS,2) is made equal to ACC divided by NPIX-OS (block 116) . This is done to normalize the abso ⁇ lute value of the differences in intensities between the two sums of pixel maps. The reason for this is that as the offset is shifted (refer to FIG. 8) , there are fewer and fewer pixels in the overlap region.
- FIG. 8 represents the left-hand and right-hand pixel memory maps stored in RAM and designated 77 and 79, respectively.
- the arrows pointing between the memory maps 77 and 79 represent the summation of the absolute differences of the intensities at the pixel locations which are now represented as addresses in the memory map that are in the current overlap region. It can be seen that as the offset increases, the overlap region gets smaller.
- the variable OS i ⁇ increased by one until the offset has reached a maximum which theoretically could be equal to the number of pixels, but, as a practical matter, is limited to about 20% of a line (about 100 pixels) for a typical 511 pixels/line camera. When this maximum is reached, the software then operates on the variables QU01 (OS,l) and QUOl (OS,2).
- X is first made equal to zero and OS is made equal to 1 (block 122) .
- X is a null counter variable and OS is the address of the second sum in the sum of differences list which is the QUOl array.
- QUOl is a two dimensional array containing the sums of differ ⁇ ences and the offset values at which they were calcu ⁇ lated. This array has now been normalized as indicated above at block 116. Referring to blocks 124 and 126, each sum in the QUOl list is compared to the adjacent sums in the list.
- FIG. 9d describes a search routine which searches for a global NULL.
- the first NULL value in the NULL list is placed in a register named GLNUL.
- the corresponding offset is placed in a register named GNOS (block 136) .
- the NULL list pointer NL is then set to the second value in the list where NL equals 2 (block 138) .
- This NULL value is compared to the previous NULL value (block 140) , and if the NULL value is less than the previous NULL value, it replaces that value in the GLNUL register and its offset replaces the value in the GNOS register (block 142) .
- NL is then incremented (block 145) and the process is repeated until all X values in the list have been tested (block 146) .
- the NULL value that survives in the GLNUL register will be the lowest value in the NULL list, and its corresponding offset in the GNOS register represents the pixel offset to the global NULL.
- the global NULL offset from the first line pair is selected (block 148) .
- distance d is defined as the offset of the global NULL divided by two (block 150) .
- an angle ⁇ is made equal to d times a proportionality constant ⁇ PIX (block 152) .
- the trigo ⁇ nometric equation which yields the range is then solved where R, which is the range, is equal to b/2 tan (90°- ⁇ ) (block 154) . This is performed for the first range calculation and the program then loops back and performs it for the second range calculation (blocks 156 and 158) .
- the range value is stored in memory (block 162) .
- Speed calculation is shown in FIG. 9f.
- the time of capture is fetched from storage for the first and second frame pairs, respectively, which represent the values T 1 and T 2 (block 164) .
- the speed is then calcu ⁇ lated as the change in the values R, to R 2 as a function of the difference in times (block 166) .
- the speed may then be stored and displayed along with the calendar date and other alphanumeric information generated by the real time clock and calendar 46 (block 168) . Once this occurs, the "command received" flag is cleared and the program returns to start (block 170) .
- FIGS. 10 and Ila through lie A second embodiment of the invention is shown in FIGS. 10 and Ila through lie.
- the hardware configura- tion for this embodiment is the same as described above in FIGS. 1, 2 and 3.
- What is different about this alter ⁇ native embodiment, termed the "edge detection" method is the way in which software present in the computer 30 controls the acquisition of video data from the cameras and processe ⁇ that data to determine the angle between the right- and left-hand cameras and the target.
- the underlying theory of the edge detection method is that a moving target will change position in the overlapping fields of view of the two narrow field-of-view cameras while the background will remain essentially the same.
- the right-hand and left-hand edges of the moving target may be determined and the angle to the target may then be calculated.
- a line of video data termed LINE 1 shows an image in the field of view of the right- and left-hand cameras.
- LINE 2 is a video image of the same field of view but taken at a later time. As the diagram shows, the video line images are nearly identical except for the apparent movement of an object which is presumed to be a moving target image.
- LINE 1 is subtracted by LINE 2, the background video information is eliminated and the edges of the moving target are seen as video pulse ⁇ that exceed predetermined positive and negative thresholds. This locates the edges of the moving target, and, armed with this information, the system can calculate the location of the center of the moving target and thus the angle to the target.
- This method is illustrated in the flow chart diagram shown in FIG. Ila through lie.
- the cameras capture left-hand and right-hand images of the target (block 174) .
- the time of capture is recorded (block 176) and the process is repeated at three different times until three line pairs are captured (blocks 178 and 180) .
- the system then subtracts the first two images pixel by pixel to form a differential line image (block 182) .
- the root mean square (rms) value of each differential line is calculated (block 184) .
- the same process then occurs for the second and third line image pair (blocks 186 and 188) .
- the system sets positive and negative thresholds for each differential line image (blocks 190 and 192) as a func- tion of the rms values.
- a "thresh- olded" differential line is obtained (block 194) .
- the system has obtained data representing a differential video line image that corresponds to the bottom waveform of FIG. 10.
- the system must then calculate the position of respective left-hand and right- hand edges. It does so by starting with the first pixel in the line and determining whether that pixel is above the positive threshold or below the negative threshold (blocks 196, 198 and 200) .
- pixel values are either given a positive 1, a negative 1, or a zero value (blocks 202, 204 and 206) . This process repeats until the last pixel on the line is reached (blocks 208 and 210) . Referring to FIG.
- the software searches for the first non-zero pixel (block 212) and this pixel number is recorded as a first edge of the target image (block 214) .
- the system searches for the last non-zero pixel (block 216) and records this pixel number as the second edge of the target image (block 218) .
- the target image center location may then be calculated (block 220) . This calculation yields a pixel offset number which may then be converted into an angle in the manner described in FIG. 9e (block 222).
- the range to the target for each differential line pair may be calculated (block 224) and, with this information, the system is able to calculate the speed (block 226) in the manner shown in FIGS. 9e and 9f.
- the edge detection method does not assume that the target image includes a point centered midway between the left-hand and right- hand cameras. In fact, the more general geometric relation shown in FIG. 6 applies. Thus, in order to determine the range R, ⁇ LH and ⁇ RH must both be known. In calculating these angles, the only approximation made is that ⁇ is directly proportional to the pixel offset which represents the midpoint between the two target edges. This will hold true when R is much greater than b.
- the angle ⁇ for both the right-hand and left- hand cameras i ⁇ calculated by determining the displace ⁇ ment of the target image center from the line of sight.
- the line of sight for both the right-hand and left-hand cameras is the pixel po ⁇ ition which corre ⁇ pond ⁇ to the center of the image line scan.
- a distance d can be obtained for both the right-hand and left-hand cameras and ⁇ will then be equal to a proportionality constant multiplied by the distance d.
- angle ⁇ is equal to 90° minus ⁇ LH and ⁇ is equal to 90° minus ⁇ RH.
- the range is, therefore, calculated according to the following expression:
- the edge detection method works best for objects travelling parallel to the baseline b across the field of view of the cameras.
- the system will also work for objects entering the field of view in a direction perpendicular to the baseline or at an obtuse angle.
- Adjustments in the calculation algorithm may be made to include a trigonometric function which account ⁇ for differences in the angular orientation of the moving target with respect to the field of view. This will be a known quantity which can be represented by a constant, since for applications such as traffic control, the moving objects travel in an undeviating straight line across the field of view.
- FIGS. 13, 14, 15 and 16 Referring to FIG.
- the video scan lines for the left-hand and right-hand camera lense ⁇ are parallel to the ba ⁇ eline.
- the fields of view of the left-hand and right-hand camera lenses are offset vertically for purposes of illustration only.
- FIG. 13 A different configuration is shown in which the cameras are mounted along a vertical baseline a predetermined distance apart and the scan lines extend perpendicular to the baseline. The cameras could also be mounted in the horizontal plane but in such a case the scan lines on the light sensitive device would be rotated 90".
- the difference in performance is illustrated in FIG. 14.
- the pixel maps consist of complete lines of video data instead of fractions of lines.
- Resolution is higher for this embodiment because a complete line of video data takes the place of a fraction of a line that can be as short as 100 pixels. There may be as many as 489 lines having at least 500 pixels per line in conventional camera systems which would employ this technique. As with the single line scan embodiment, an offset distance d will be determined which represents an offset from the center of the field of view to the estimated location of the target image. Although resolution of the global null for this embodiment is better, processing time is slower because of the need to scan an entire frame consisting of as many as 489 lines, and the memory for storing the video data must be made larger in order to accommodate the additional information.
- FIG. 15 shows the sy ⁇ tem configured with an upper camera 230 and a lower camera 232 mounted along a vertical ba ⁇ eline and separated by a distance b.
- the upper camera 230 includes a field of view that extends from LI UPR to L489 UPR.
- the lower camera includes a field of view that extends from LI LWR to L489 LWR.
- the drawing show ⁇ the overlap region at a range R j .
- the dashed lines extending from the upper and lower cameras 230 and 232, respectively, indicate the position of an arbitrarily chosen scanning line, termed herein ⁇ after a "template" line.
- L TEMP L TEMP
- the template line is an image scan line containing at least a portion of the target image at any arbitrary range.
- the pixel intensitie ⁇ from all lines (the entire frame) of the upper camera are compared to the template line. (Only the pixel intensities from the template line of the lower camera are mapped in com ⁇ puter memory.)
- the upper camera output is then ⁇ hifted and compared, line by line, to the ⁇ ingle template line from the lower camera frame map until the image line locations correlate. This occurs at a total shift of 2d lines. This relationship is illustrated as shown in FIG.
- the offset is made equal to the template line (block 234) .
- ACC is set equal to 0 (block 236) and PIX is set equal to l (block 238) .
- ACC is then made equal to the absolute difference between an arbitrary line in the upper field of view and the template line (block 240) .
- the variable PIX is incre- mented and the process is repeated until PIX equals NPIX (blocks 242 and 244) .
- the variable PIX is an array pointer variable common to both arrays and thus points to the like numbered pixel in both arrays.
- the line pointer for the upper map is the variable OS whereas the line pointer for the lower array is a constant (TEMP) .
- QUOl is a two dimensional array containing the sums of differ ⁇ ences and the offset values at which they were calcu ⁇ lated. Note that the sums of differences in this case do not need to be normalized since all are made up of the same number of differences.
- the proces ⁇ proceeds until the maximum possible offset is reached and all values of QUOl have been calculated (blocks 246, 248, 250 and 252).
- the program then goes to FIG. 9C to calculate the nulls and to find the global null. The only difference will be in the value of the proportionality constant ⁇ PIX which is now made a function of the distance between the centers of adjacent lines instead of between the centers of adjacent pixels.
- FIG. 17 shows a schematic view of such an arrangement.
- a left-hand lens 260 and a right-hand lens 262 focus light through a pair of reflec ⁇ tive prisms 264 and 266, respectively.
- These prisms direct light onto a stacked prism arrangement 268 having an upper and lower prism 268a and 268b, respectively.
- the upper and lower prisms 268a and 268b reflect the light onto different regions of a single CCD line or area camera chip 270.
- FIG. 17 is useful primarily for the single line scanning method of FIG. 7 and the edge detection method of FIG. 10.
- a configuration appropriate to the multiple line scanning method is shown in FIG. 19.
- the lenses are mounted side-by-side but the line scanning is conducted in a plane perpendicular to the plane of the drawing.
- Light from the object enters a left-hand lens 272 and is reflected off of mirrors 274 and 276 onto a
- CCD 278 A small area in the top of the field of view of the lens 272 is chosen and dedicated to the left-hand lens. This area is shown between the arrows and repre ⁇ sents the template line scanning area.
- the rest of the CCD 278 may be scanned by light from the right-hand lens 280 reflected from mirror 282 and prism 284.
- a single camera may be constructed which includes two lenses mounted side-by-side or one on top of the other along a common baseline and the light from both lenses may be imaged onto a single light sensitive device.
- the signal processing for at least a portion of the "baseline perpendicular" multiple line scanning embodiment may also be employed as shown in FIG. 18.
- a digital preprocessor replaces the functions performed by the computer in FIG. 16.
- the computer hardware complement of FIG. 3 is altered so that frame grabbers 32 and 34 are no longer needed.
- This digital preprocessor may be constructed from commonly available off-the-shelf components and significantly reduces the computational requirements of the computer. Range measurements can, therefore, be made more frequently thus improving the accuracy of the speed computation. Also, less RAM is required in the computer.
- the sums of differences calculations described in the flow chart of FIG. 16 are performed by the preprocessor of FIG. 18 and the results are transferred to a computer interface 318.
- the upper camera termed the template camera 300, transfers video to an eight bit analog-to- digital converter 302.
- the output of the eight bit ADC is provided to a 511 (pixel) by eight bit memory 304.
- the lower or frame camera 306 transfers video to its eight bit ADC 308.
- Both cameras can be RS 170 standard video cameras synchronized by an RS 170 sync generator 310.
- Clocking and timing circuits 312 cause digital video data representing the template line to be stored in the memory and clocked out later and summed with the output of ADC 308 in a sums of differences computation device termed a complementor/adder/accumulator 314.
- Computation device 314 is programmed to operate in accordance with the program outlined on FIG.
- the linear regression method may be implemented by storing the measured ranges R n and their associated times T n in a two dimensional array in computer memory. Speed may then be determined based upon the magnitude and slope of the line that is fitted to the group of cartesian coordinates. If the algebraic sign of the slope is positive, the target is receding. If negative, the target is approaching. A confidence level which is based upon the square of the range coordinates may also be calculated.
- an array of range measurements taken at a plurality of times is shown in chart form.
- the range measurements are shown in column 1 and comprise R-, R 2 . . . R n taken at the times shown in column 2 of Tl, T. . . . T n .
- the number of the measure ⁇ ment is designated by the row number and extends from 1 to RCMAX.
- the row number indicates an address in memory at which the two dimensional column array is stored.
- FIG. 20 shows a general flow chart diagram of the process of the invention using a linear regression method to calculate the velocity of the moving target.
- This chart is similar to the chart of FIG. 4.
- Blocks 62, 64, 66, 68 and 70 are the identical proce ⁇ s steps shown in FIG. 4. The difference from the proces ⁇ illustrated in FIG. 4 begins at block 330. At this point the process of blocks 62 through 70 is repeated at later times to obtain a plurality of range and time measurements R n at times Tn. At least three such measurements are obtained
- Blocks 332, 334 and 336 illustrate the linear regression calculation that yields information regarding the speed of the target.
- a linear regres ⁇ sion calculation is undertaken to fit the best straight line to the cartesian data points represented by range and time.
- the ⁇ lope of thi ⁇ line is calcu ⁇ lated.
- the speed of the target is directly proportional to this slope.
- a confidence level may be calculated.
- the results of the calculations of blocks 332, 334 and 336 are displayed at block 338.
- variables X and Y are initialized.
- a variable to be entitled "ROW" is set at 1 in block 342.
- a linear regre ⁇ sion straight line fit for the range and time coordinates represented by variables X and Y is calculated in block 344 in the classical manner.
- the program loop ⁇ back to block 344 a ⁇ long a ⁇ the variable ROW is not equal to RCMAX plus 1.
- the program next calculates R based upon the sums of variables X and Y (block 350) .
- a confidence level represented by R 2 is calcu- lated.
- the calculation in block 354 represents the slope of the straight line which represents the linear fit to the cartesian coordinates X, Y.
- the speed is calculated in block 356 by simply multiplying the slope by a constant.
- the speed and confidence level are displayed.
- correlation between pixel maps may be done at every nth pixel in the overlap region instead of at each pixel offset.
- the minimum overlap region is about 100 pixels and can extend to 511 pixels.
- the selected value of n represents a trade off between the maximum spatial frequency resolvable with the algorithm and the time required to perform the algorithm. For a given processor, this is a trade off between the accuracy of each range measurement and the number of range measure ⁇ ments that can be made in a given time.
- the linear regression technique described above can be used.
- variable PIX plus 1 in block 110 of FIG. 9b would be changed to PIX plus PIXSTEP where PIXSTEP equals n.
- the right hand side of the equation would change to PIXSTEP *ACC/NPIX minus OS.
- block 242 the variable PIX plus 1 would change to PIX plus PIXSTEP.
- This technique provides a more course resolu ⁇ tion of the target but is much faster since only every nth pixel must be correlated. This permits a larger number of range measurements which would yield greater range accuracy where a technique such as linear regression is u ⁇ ed.
- the invention has been described with reference to detection systems for detecting the speed of moving vehicles it should be under ⁇ tood that the inven- tion described herein has much broader application, and in fact may be used to detect the range to a stationary object, the speed of any moving object and/or relative motion between moving or ⁇ tationary object ⁇ .
- the invention may be incorporated in a robotics manufacturing or monitoring sy ⁇ tem for monitoring or operating upon object ⁇ moving along an a ⁇ embly line.
- Another important application i ⁇ a ranging device u ⁇ ed in conjunction with a weapon ⁇ system for acquiring and tracking a target.
- Yet another application is a spotting system used to detect camouflaged objects which may be in motion against a ⁇ tatic background.
- edge detection embodiment di ⁇ closed above is especially useful for thi ⁇ purpose.
- Other possible uses and applications will be apparent to those skilled in the art.
- the terms and expressions which have been employed in the foregoing specification are used therein as terms of description and not of limitation, and there is no intention, in the use of such terms and expres ⁇ sions, of excluding equivalents of the features shown and described or portions thereof, it being recognized that the scope of the invention is defined and limited only by the claims which follow.
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Power Engineering (AREA)
- Measurement Of Optical Distance (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US1995/000819 WO1996022537A1 (en) | 1995-01-18 | 1995-01-18 | Optical range and speed detection system |
Publications (2)
Publication Number | Publication Date |
---|---|
EP0877944A1 true EP0877944A1 (de) | 1998-11-18 |
EP0877944A4 EP0877944A4 (de) | 1999-04-21 |
Family
ID=22248540
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP95910109A Withdrawn EP0877944A4 (de) | 1995-01-18 | 1995-01-18 | Optisches entfernungs-und geschwindigkeitsmessystem |
Country Status (5)
Country | Link |
---|---|
EP (1) | EP0877944A4 (de) |
JP (1) | JPH10512953A (de) |
AU (1) | AU690230B2 (de) |
CA (1) | CA2209610C (de) |
WO (1) | WO1996022537A1 (de) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BE1014144A3 (fr) * | 2001-04-27 | 2003-05-06 | Visee Christian | Methode et dispositif pour evaluer un parametre d'un objet en mouvement. |
IT1399162B1 (it) * | 2010-04-01 | 2013-04-11 | Consorzio Naz Interuniversitario Per Le Telecomunicazioni Cnit | Sistema di monitoraggio |
CN102393972B (zh) * | 2011-07-21 | 2014-04-23 | 深圳市爱森斯信息技术有限公司 | 闸机通道检测***及其检测方法 |
RU2524441C2 (ru) * | 2012-05-02 | 2014-07-27 | Федеральное государственное бюджетное образовательное учреждение высшего профессионального образования "Самарский государственный технический университет" | Устройство для измерения вектора скорости движения изображения объекта со случайным распределением яркостей |
US9131295B2 (en) | 2012-08-07 | 2015-09-08 | Microsoft Technology Licensing, Llc | Multi-microphone audio source separation based on combined statistical angle distributions |
US9269146B2 (en) | 2012-08-23 | 2016-02-23 | Microsoft Technology Licensing, Llc | Target object angle determination using multiple cameras |
JP2015064324A (ja) * | 2013-09-26 | 2015-04-09 | 株式会社リコー | 移動部材検出装置 |
CN107395972B (zh) * | 2017-07-31 | 2020-03-06 | 华勤通讯技术有限公司 | 一种快速移动对象的拍摄方法及终端 |
CN113419075B (zh) * | 2021-05-31 | 2022-08-16 | 广州忘平信息科技有限公司 | 基于双目视觉的船舶测速方法、***、装置和介质 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0035794A1 (de) * | 1980-03-12 | 1981-09-16 | Siemens Aktiengesellschaft | Anordnung zur optoelektronischen Entfernungsmessung |
US4320302A (en) * | 1978-09-05 | 1982-03-16 | Siemens Aktiengesellschaft | Method for the automatic or semiautomatic distance measurement and focusing of the image of an object onto a focal plane |
US4601053A (en) * | 1983-11-21 | 1986-07-15 | Grumman Aerospace Corporation | Automatic TV ranging system |
DE3642196A1 (de) * | 1986-12-10 | 1988-06-23 | Mel Mikro Elektronik Gmbh | Optoelektronische kollisionsschutzvorrichtung fuer fahrzeuge |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3788201A (en) * | 1970-12-14 | 1974-01-29 | F Abell | Method for establishing vehicle identification, speed and conditions of visibility |
US4173010A (en) * | 1975-05-01 | 1979-10-30 | Hoffmann Anton R | Traffic sign and improved system for recording vehicle speed |
US4135817A (en) * | 1978-03-09 | 1979-01-23 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | Apparatus for measuring an aircraft's speed and height |
US4257703A (en) * | 1979-03-15 | 1981-03-24 | The Bendix Corporation | Collision avoidance using optical pattern growth rate |
US4433325A (en) * | 1980-09-30 | 1984-02-21 | Omron Tateisi Electronics, Co. | Optical vehicle detection system |
US4495589A (en) * | 1982-09-20 | 1985-01-22 | Crane Co. | Aircraft ground velocity determination system |
US4671650A (en) * | 1982-09-20 | 1987-06-09 | Crane Co. (Hydro-Aire Division) | Apparatus and method for determining aircraft position and velocity |
DE3516664C2 (de) * | 1985-05-09 | 1987-03-19 | Honeywell Regelsysteme GmbH, 6050 Offenbach | Optische Anordnung zur passiven Entfernungsbestimmung |
US4847772A (en) * | 1987-02-17 | 1989-07-11 | Regents Of The University Of Minnesota | Vehicle detection through image processing for traffic surveillance and control |
IL86202A (en) * | 1988-04-27 | 1992-01-15 | Driver Safety Systems Ltd | Traffic safety monitoring apparatus |
US4969735A (en) * | 1989-03-07 | 1990-11-13 | Sperry Marine Inc. | Passive range finding apparatus utilizing television sensors |
JP2583641B2 (ja) * | 1990-05-18 | 1997-02-19 | 日産自動車株式会社 | 走行制御方法 |
JPH0812072B2 (ja) * | 1990-06-13 | 1996-02-07 | 三菱電機株式会社 | 距離測定装置 |
US5161107A (en) * | 1990-10-25 | 1992-11-03 | Mestech Creation Corporation | Traffic surveillance system |
JP3167752B2 (ja) * | 1991-10-22 | 2001-05-21 | 富士重工業株式会社 | 車輌用距離検出装置 |
-
1995
- 1995-01-18 AU AU18329/95A patent/AU690230B2/en not_active Ceased
- 1995-01-18 JP JP8522222A patent/JPH10512953A/ja active Pending
- 1995-01-18 CA CA002209610A patent/CA2209610C/en not_active Expired - Fee Related
- 1995-01-18 WO PCT/US1995/000819 patent/WO1996022537A1/en not_active Application Discontinuation
- 1995-01-18 EP EP95910109A patent/EP0877944A4/de not_active Withdrawn
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4320302A (en) * | 1978-09-05 | 1982-03-16 | Siemens Aktiengesellschaft | Method for the automatic or semiautomatic distance measurement and focusing of the image of an object onto a focal plane |
EP0035794A1 (de) * | 1980-03-12 | 1981-09-16 | Siemens Aktiengesellschaft | Anordnung zur optoelektronischen Entfernungsmessung |
US4601053A (en) * | 1983-11-21 | 1986-07-15 | Grumman Aerospace Corporation | Automatic TV ranging system |
DE3642196A1 (de) * | 1986-12-10 | 1988-06-23 | Mel Mikro Elektronik Gmbh | Optoelektronische kollisionsschutzvorrichtung fuer fahrzeuge |
Non-Patent Citations (1)
Title |
---|
See also references of WO9622537A1 * |
Also Published As
Publication number | Publication date |
---|---|
AU1832995A (en) | 1996-08-07 |
EP0877944A4 (de) | 1999-04-21 |
JPH10512953A (ja) | 1998-12-08 |
WO1996022537A1 (en) | 1996-07-25 |
AU690230B2 (en) | 1998-04-23 |
CA2209610A1 (en) | 1996-07-25 |
CA2209610C (en) | 2001-03-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5586063A (en) | Optical range and speed detection system | |
CN1302290C (zh) | 激光测位装置及方法 | |
US5589942A (en) | Real time three dimensional sensing system | |
US6042050A (en) | Synthetic discriminant function automatic target recognition system augmented by LADAR | |
US4802757A (en) | System for determining the attitude of a moving imaging sensor platform or the like | |
US11733043B2 (en) | Automatic locating of target marks | |
JP3953103B2 (ja) | 標的マークの位置を敏速に検出するための方法及び装置 | |
US20040247157A1 (en) | Method for preparing image information | |
US5559322A (en) | Imaging optical tracker | |
CN1813201A (zh) | 物体探测***和探测物体的方法 | |
US6675121B1 (en) | Velocity measuring system | |
US7375826B1 (en) | High speed three-dimensional laser scanner with real time processing | |
US5424823A (en) | System for identifying flat orthogonal objects using reflected energy signals | |
GB2166920A (en) | Measuring angular deviation | |
JPH09226490A (ja) | 横断物の検出装置 | |
EP0877944A1 (de) | Optisches entfernungs-und geschwindigkeitsmessystem | |
Hattori et al. | Accurate rangefinder with laser pattern shifting | |
US5309159A (en) | Method and system for moving object detection | |
El-Hakim et al. | Integration of range and intensity data to improve vision-based three-dimensional measurements | |
US4993830A (en) | Depth and distance measuring system | |
CN113256483A (zh) | 用于目标识别的点云数据去抖动 | |
CN116027342A (zh) | 一种激光测距方法、***、激光雷达及雷达制品 | |
Hebel et al. | LiDAR-supported navigation of UAVs over urban areas | |
CN114509762A (zh) | 一种数据处理方法、装置、设备及介质 | |
US20070162248A1 (en) | Optical system for detecting intruders |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 19970723 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LI LU MC NL PT SE |
|
A4 | Supplementary search report drawn up and despatched |
Effective date: 19990308 |
|
AK | Designated contracting states |
Kind code of ref document: A4 Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LI LU MC NL PT SE |
|
17Q | First examination report despatched |
Effective date: 20000915 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION HAS BEEN WITHDRAWN |
|
18W | Application withdrawn |
Withdrawal date: 20020808 |