WO2012002069A1 - 形状抽出方法及び装置、並びに寸法測定装置及び距離測定装置 - Google Patents
形状抽出方法及び装置、並びに寸法測定装置及び距離測定装置 Download PDFInfo
- Publication number
- WO2012002069A1 WO2012002069A1 PCT/JP2011/061800 JP2011061800W WO2012002069A1 WO 2012002069 A1 WO2012002069 A1 WO 2012002069A1 JP 2011061800 W JP2011061800 W JP 2011061800W WO 2012002069 A1 WO2012002069 A1 WO 2012002069A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- individual
- shape
- unit
- line
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B5/00—Measuring arrangements characterised by the use of mechanical techniques
- G01B5/0035—Measuring of dimensions of trees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30161—Wood; Lumber
Definitions
- the present invention relates to a method and apparatus for extracting the shape of a specific object from a captured image, and an apparatus for measuring the size of the extracted object and the distance to the object.
- This dimension measuring device identifies an image of an object from the screen based on feature values such as color and shape, and extracts the shape of the identified object image. And the dimension of a target object is calculated based on the shape of the extracted target object image.
- the shape of the object image is extracted, the image is converted into a line image represented by a contour line, and is performed based on the contour line included in the line image.
- the lung region extraction apparatus described in Patent Document 1 removes a contour line that touches the boundary of an image by positioning the object image in the center of the screen when extracting the object image.
- the shape of the object image is accurately extracted by not extracting anything other than the outline of the object image.
- the line image represented by the contour line includes a partial image and a pattern line (pseudo contour line) in the target image in addition to the contour line (outline contour line) representing the outer shape of the object image. . Therefore, in order to accurately extract the shape of the object image, it is necessary to correctly extract the outline contour line from the line image.
- the pseudo contour line in contact with the boundary of the image is removed, but other pseudo contour lines, for example, the pseudo contour line inside the outer contour line or the outer contour line
- the pseudo contour line that is outside the border of the image and is not in contact with the boundary of the image is not removed. For this reason, many pseudo contour lines are extracted in addition to the external contour lines, and the shape of the object image cannot be accurately extracted.
- An object of the present invention is to provide a shape extraction method and apparatus, a dimension measurement apparatus, and a distance measurement apparatus that can accurately extract the shape of an object image.
- the shape extraction apparatus of the present invention includes an individual image region extraction unit, an individual line image generation unit, a removal unit, an estimated shape determination unit, a determination unit, an object shape identification unit, It has.
- the individual image region extraction unit identifies a plurality of individual images estimated as objects having a specific shape from the screen, and extracts an individual image region where the individual images exist from the screen.
- the individual line image generation unit sequentially selects each individual image area, converts each image including the individual image in the selected individual image area into a line image, and generates an individual line image.
- the removing unit has a center at substantially the same position as the center of the individual line image from the contour lines included in the individual line images, and the vertical and horizontal lengths of the individual line images are substantially the same as those of the individual line images.
- a contour line that is less than half and / or a contour line that exists on a straight line connecting the center point of the contour line itself and the center point of the individual line image is removed as an unnecessary contour line.
- the estimated shape determining unit determines an estimated shape having an optimum size of the specific shape so as to correspond to the corrected line image from which the unnecessary contour line is removed.
- the determination unit evaluates the degree of overlap between the estimated shape and the corrected line image to determine whether the estimated shape is appropriate.
- the object shape specifying unit specifies the estimated shape as the shape of the object extracted from the screen when it is determined that the estimated shape is appropriate.
- Examples of the object include logs and boxes.
- Examples of the method for identifying the object include identification by shape, identification by color, and the like.
- the removal unit also removes the contour line of the corrected line image related to the object whose shape has been previously specified when the individual image areas overlap.
- the removing unit also removes a contour line in contact with a boundary of the individual image area.
- the estimated shape determination unit divides the corrected line image into a plurality of regions, selects a contour line from each of the plurality of regions, and determines the estimated shape based on the selected plurality of contour lines. Is preferred.
- the removing unit removes the contour line outside the estimated shape.
- the individual line image generation unit preferably includes a plurality of converted image extraction filters having different threshold values for extracting an image to be converted into a line image from images included in the selected individual image region. .
- the individual line image generation unit generates an individual line image by converting an image extracted using any one of the plurality of converted image extraction filters into a line image.
- the individual line image generation unit changes the converted image extraction filter and generates a new individual line image.
- the individual line image generation unit includes a plurality of converted image extraction filters having different threshold values for extracting an image to be converted into a line image from images included in the selected individual image region. .
- the individual line image generation unit converts the images extracted using the converted image extraction filters into line images, and generates a plurality of individual line images for one individual image region.
- the estimated shape determination unit determines a plurality of estimated shapes using each of the plurality of individual line images.
- the determination unit determines each of the plurality of estimated shapes.
- the object shape specifying unit specifies, as the shape of the object, one of the plurality of estimated shapes that has the highest evaluation in the determination unit.
- the individual line image generation unit has a plurality of thresholds for binarizing an image included in the selected individual image region to generate the individual line image.
- the individual line image generation unit binarizes an image included in the selected individual image region based on any of the plurality of threshold values, and generates the individual line image.
- the individual line image generation unit changes the threshold value and generates a new individual line image.
- the individual line image generation unit has a plurality of thresholds for binarizing an image included in the selected individual image region to generate the individual line image.
- the individual line image generation unit binarizes an image included in the selected individual image region based on each of the plurality of threshold values, and generates a plurality of individual line images for one individual image region.
- the estimated shape determination unit determines a plurality of estimated shapes using each of the plurality of individual line images.
- the determination unit determines each of the plurality of estimated shapes.
- the object shape specifying unit specifies, as the shape of the object, one of the plurality of estimated shapes that has the highest evaluation in the determination unit.
- the threshold calculation unit binarizes an image included in the selected individual image region and calculates a threshold for generating the individual line image.
- the individual line image generation unit binarizes an image included in the selected individual image region based on the threshold value calculated by the threshold value calculation unit, and generates the individual line image.
- the threshold value calculation unit calculates a new threshold value.
- the individual line image generation unit generates a new individual line image based on the new threshold value.
- the threshold calculation unit calculates a plurality of thresholds for binarizing an image included in the selected individual image region to generate the individual line image.
- the individual line image generation unit binarizes an image included in the selected individual image region based on each of the plurality of threshold values, and generates a plurality of individual line images for one individual image region.
- the estimated shape determination unit determines a plurality of estimated shapes using each of the plurality of individual line images.
- the determination unit determines each of the plurality of estimated shapes.
- the object shape specifying unit specifies, as the shape of the object, one of the plurality of estimated shapes that has the highest evaluation in the determination unit.
- the dimension measuring apparatus of the present invention includes an individual image region extracting unit, an individual line image generating unit, a removing unit, an estimated shape determining unit, a determining unit, an object shape specifying unit, a distance measuring unit, and a dimension measuring unit. And an actual size calculation unit.
- the individual image region extraction unit identifies a plurality of individual images estimated as objects having a specific shape from the screen, and extracts an individual image region where the individual images exist from the screen.
- the individual line image generation unit sequentially selects each individual image area, converts each image including the individual image in the selected individual image area into a line image, and generates an individual line image.
- the removing unit has a center at substantially the same position as the center of the individual line image from the contour lines included in the individual line images, and the vertical and horizontal lengths of the individual line images are substantially the same as those of the individual line images.
- a contour line that is less than half and / or a contour line that exists on a straight line connecting the center point of the contour line itself and the center point of the individual line image is removed as an unnecessary contour line.
- the estimated shape determining unit determines an estimated shape having an optimum size of the specific shape so as to correspond to the corrected line image from which the unnecessary contour line is removed.
- the determination unit evaluates the degree of overlap between the estimated shape and the corrected line image to determine whether the estimated shape is appropriate.
- the object shape specifying unit specifies the estimated shape as the shape of the object extracted from the screen when it is determined that the estimated shape is appropriate.
- the distance measuring unit measures a distance to the object when photographing the object.
- the dimension measuring unit measures a dimension of the object shape specified by the object shape specifying part.
- the actual dimension calculation unit calculates the actual dimension of the object based on the distance measured by the distance measurement unit and the dimension measured by the dimension measurement unit.
- the magnification input unit inputs a photographing magnification when photographing the object.
- the actual dimension calculation unit is configured to determine the actual dimension of the object based on the distance measured by the distance measurement unit, the dimension measured by the dimension measurement unit, and the photographing magnification input from the magnification input unit. Is calculated.
- the distance measuring device of the present invention includes an individual image region extracting unit, an individual line image generating unit, a removing unit, an estimated shape determining unit, a determining unit, an object shape specifying unit, an actual dimension input unit, dimensions
- a measurement unit and a distance calculation unit are provided.
- the individual image region extraction unit identifies a plurality of individual images estimated as objects having a specific shape from the screen, and extracts an individual image region where the individual images exist from the screen.
- the individual line image generation unit sequentially selects each individual image area, converts each image including the individual image in the selected individual image area into a line image, and generates an individual line image.
- the removing unit has a center at substantially the same position as the center of the individual line image from the contour lines included in the individual line images, and the vertical and horizontal lengths of the individual line images are substantially the same as those of the individual line images.
- a contour line that is less than half and / or a contour line that exists on a straight line connecting the center point of the contour line itself and the center point of the individual line image is removed as an unnecessary contour line.
- the estimated shape determining unit determines an estimated shape having an optimum size of the specific shape so as to correspond to the corrected line image from which the unnecessary contour line is removed.
- the determination unit evaluates the degree of overlap between the estimated shape and the corrected line image to determine whether the estimated shape is appropriate.
- the object shape specifying unit specifies the estimated shape as the shape of the object extracted from the screen when it is determined that the estimated shape is appropriate.
- the actual dimension input unit inputs an actual dimension of the object.
- the dimension measuring unit measures a dimension of the object shape specified by the object shape specifying part.
- the distance calculation unit calculates a distance to the object when the object is photographed based on the actual dimension input by the actual dimension input unit and the dimension measured by the dimension measurement unit. .
- magnification input unit inputs a photographing magnification when photographing the object.
- the distance calculation unit is based on the actual dimension input by the actual dimension input unit, the dimension measured by the dimension measurement unit, and the photographing magnification input by the magnification input unit. Calculate the distance.
- the shape extraction method of the present invention includes an individual image region extraction step, an individual line image generation step, a removal step, an estimated shape determination step, a determination step, and an object shape identification step.
- an individual image region extraction step an individual image estimated to be an object having a specific shape is identified from the screen, and an individual image region in which each individual image exists is extracted from the screen.
- the individual line image generation step the individual image areas are selected in order, and an image including the individual image in the selected individual image area is converted into a line image to generate an individual line image.
- the removal step has a center at substantially the same position as the center of the individual line image from the contour lines included in each individual line image, and the vertical and horizontal lengths are substantially the same as those of the individual line image.
- a contour line that is less than half and / or a contour line that exists on a straight line connecting the center point of the contour line itself and the center point of the individual line image is removed as an unnecessary contour line.
- the estimated shape determining step determines an estimated shape having the specific shape as an optimum size so as to correspond to the corrected line image from which the unnecessary contour line is removed. In the determination step, the degree of overlap between the estimated shape and the corrected line image is evaluated to determine whether the estimated shape is appropriate.
- the object shape specifying step specifies the estimated shape as the shape of the object extracted from the screen when it is determined that the estimated shape is appropriate.
- the contour line has a center at substantially the same position as the center of the individual line image, and the length in the vertical direction and the horizontal direction is approximately half or less of the individual line image.
- / or the contour line existing on the straight line connecting the center point of the contour line and the center point of the individual line image is removed.
- an estimated shape having an optimum size of the specific shape is determined so as to correspond to the corrected line image from which the unnecessary contour line is removed, and the shape of the object is specified using the estimated shape. Thereby, the shape of a target object can be extracted correctly.
- the contour line related to the object whose shape has been specified is also deleted, so that the shape of the object can be specified more accurately.
- the digital camera 10 is provided with a lens barrel 11 on the front surface of a camera body 10a.
- An imaging lens 12 is incorporated in the lens barrel 11.
- the lens barrel 11 is housed inside the camera body 10a when the digital camera 10 is in a power-off state, and protrudes from the front surface of the camera body 10a and is set at a wide position when the power is on.
- a zoom lens having a variable power lens 12a and a focus lens 12b is used as the imaging lens 12.
- a flash light emitting unit 15 is provided on the front surface of the camera body 10a.
- the flash light emitting unit 15 is operated by the CPU 40 (see FIG. 3), and irradiates flash light toward the subject.
- a power button 17, a release button 18, and the like are provided on the upper surface of the camera body 10a.
- the digital camera 10 is provided with a shooting mode for shooting a still image, a playback mode for playing back and displaying the shot image, and a measurement mode for shooting an object and measuring its actual size.
- the measurement mode is provided with a plurality of measurement modes (for example, a log measurement mode, a box measurement mode, etc.) according to the type of the object.
- the release button 18 has a two-stage push structure. In the shooting mode and the measurement mode, when the release button 18 is lightly pressed (half pressed), shooting preparation operations such as exposure adjustment and focus adjustment are performed. When further pressed (fully pressed) in this state, a photographing operation is performed.
- an LCD 20 on which images and various setting conditions are displayed, a menu key 21 for setting shooting conditions and selecting a mode, and a magnification of the imaging lens 12 are changed.
- a zoom button 22 is provided. By operating the zoom button 22, the zoom lens 12a is moved between the wide-angle position and the telephoto position, and the image is enlarged or reduced.
- a color image sensor for example, a CCD 31 on which subject light transmitted through the imaging lens 12 enters is arranged.
- the CCD 31 photoelectrically converts an object image to generate a time-series three-color signal and sends it to a correlated double sampling circuit (CDS) 32.
- a TG (timing generator) 38 is controlled by the CPU 40 and generates a driving timing signal to drive the CCD 31.
- the three-color signal generated from the CCD 31 is converted into a three-color signal that accurately corresponds to the amount of charge accumulated in each cell of the CCD 31 by removing noise components by the CDS 32.
- the three-color signal output from the CDS 32 is amplified by an amplifier (AMP) 33 according to the ISO sensitivity and sent to the A / D converter 34.
- AMP amplifier
- the A / D converter 34 digitally converts the three-color signal into three-color image data (hereinafter simply referred to as image data).
- the zoom motor 39a and the focus motor 39b are driven by the CPU 40 via a motor driver (not shown), and move the zoom lens 12a and the focus lens 12b to perform zooming and focusing.
- the CPU 40 is connected to an EEPROM (not shown) in which various control programs and setting information are recorded.
- the CPU 40 reads a program in the EEPROM and controls each unit.
- the image input controller 36 sends the image data from the A / D converter 34 to the video memory 41 during display of the through image, and sends it to the buffer memory 42 during shooting.
- live view shooting live view image data with a low resolution is temporarily recorded in the video memory 41.
- the through image data is sent to the LCD driver 43 via the data bus 37 and the through image is displayed on the LCD 20.
- the captured high-resolution image data is temporarily recorded in the buffer memory 42.
- the image data read from the buffer memory 42 is recorded on the memory card 45 by the memory controller 44.
- the CPU 40 records a focal length representing a zoom magnification (photographing magnification) when the release button 18 is fully pressed on the memory card 45 via the memory controller 44.
- the photometry / ranging unit 47 detects the brightness of the subject and the distance to the subject based on the image data of the through image when the release button 18 is half-pressed, and the white balance correction amount, And the lens focus position is determined.
- the photometry / ranging unit 47 operates in a predetermined cycle while displaying a through image.
- the subject distance can be measured by well-known phase difference detection or the like.
- the photometry / ranging unit 47 sequentially transmits the luminance and subject distance detection results to the CPU 40.
- the CPU 40 moves the focus lens 12 b based on the lens focus position from the photometry / ranging unit 47. Further, the CPU 40 controls the shutter speed (charge accumulation time) of the CCD 31 and the operation of the flash light emitting unit 15 based on the luminance from the photometry / ranging unit 47.
- the battery 48 is housed inside the digital camera 10 and supplies power to each unit such as the LCD 20 and the CPU 40.
- the power supply to each unit is controlled by a control circuit (not shown) of the battery 48.
- the individual image area extraction unit 51 reads the captured image 52 in the buffer memory 42. Then, as shown in FIG. 4A, the individual image region extraction unit 51 has an individual shape (an ellipse that is a cross-sectional shape of a log in the log measurement mode and a quadrangle in the box measurement mode) that is predetermined in the captured image 52. Identify the image.
- the log measurement mode is selected, and the captured image 52 includes twelve log images 54A to 54L.
- the individual image area extraction unit 51 extracts the individual image areas 53A to 53L where the individual images of the respective objects (logs) in the captured image 52 exist, as shown in FIG. 4B, based on the identification result. In FIG. 4A and FIG. 4B, illustration of fine lines is omitted.
- the extracted individual image area is analyzed to discriminate between a log image and an image of another object according to the color of the image, and includes an image that does not include a log image, for example, a circular image similar to a log. Individual image areas may be excluded first.
- the individual line image generation unit 61 selects one of the individual image regions 53A to 53L extracted by the individual image region extraction unit 51 (for example, the individual image region 53A). .
- the individual line image generation unit 61 is provided with a converted image extraction filter 63 that extracts an image to be converted into a line image from an image in the selected individual image region.
- the converted image extraction filter 63 performs numerical analysis on the color change in the selected individual image region, and extracts an image having a color change larger than a preset extraction threshold.
- the individual line image generation unit 61 converts the image extracted by the converted image extraction filter 63 into a line image and generates an individual line image 62.
- the individual line image generation unit 61 selects the individual image area 53A, the individual image area 53B, the individual image area 53C,..., The individual image area 53K, and the individual image area 53L in this order.
- a line image 62 is generated.
- the removal unit 66 has a center at substantially the same position as the center of the individual line image 62 among the contour lines included in the individual line image 62, and the lengths in the vertical direction and the horizontal direction are individual lines.
- a contour near the center 71 (contour included in the region surrounded by a dotted line in FIG. 5B), a boundary contour 72 in contact with the boundary of the individual image region 53A, and the contour itself
- the peripheral outline 73 on the straight line (two-dot chain line in FIG. 5B) connecting the center point CP1 which is the center (center of gravity) and the center point CP2 of the individual line image 62 is removed.
- center point CP1 of the contour line itself is an average value point of the X and Y coordinates on the pixel coordinates of one contour line. Further, the central coordinate point of the straight line connecting both ends of one contour line may be set as the center point CP1 of the contour itself. Further, the region for determining the center vicinity contour line 71 can be changed as appropriate.
- the estimated shape determination unit 67 (see FIG. 3) performs elliptic fitting on the corrected line image 64 from which the contour lines 71 to 73 are removed by the removal unit 66 by a known fitting method. Do. Then, the estimated shape determining unit 67 determines an estimated ellipse shape 76 having an ellipse (specific shape) as an optimum size so as to correspond to the corrected line image 64. In the box measurement mode, box-shaped fitting is performed to determine the estimated box shape.
- the estimated shape determination unit 81 evaluates the estimated ellipse shape 76 and determines its suitability.
- the estimated shape determination unit 81 detects the total number of pixels in the overlapping portion of the remaining outline 75 in the corrected line image 64 and the estimated elliptical shape 76, and the total number of pixels exceeds a preset threshold value. When the total number of pixels is equal to or less than the threshold value, it is determined that the estimated elliptical shape 76 is not appropriate. Note that the threshold may be changed and may be determined by the user.
- the removing unit 66 removes the contour line outside the estimated elliptical shape 76. Then, the estimated ellipse shape 76 is determined and determined again, and this is repeated until the estimated ellipse shape 76 is evaluated as appropriate.
- the object shape specifying unit 83 specifies the estimated elliptical shape 76 as a log shape.
- the dimension measuring unit 84 measures the dimension of the estimated oval shape 76 specified by the object shape specifying unit 83 (the size of the log on the captured image). Note that the dimension measuring unit 84 measures the height and width of the estimated elliptical shape 76.
- the actual dimension calculation unit 85 is based on the dimensions measured by the dimension measurement unit 84, the distance to the log detected by the photometry / ranging unit 47, and the focal length (zoom magnification) of the photographing lens 12. Calculate the actual dimensions.
- the calculated actual size data of the log is recorded in the memory card 45.
- the removing unit 66 in addition to each of the contour lines 71 to 73, is a contour line on the left side or the lower side in the individual line image 62, and the determined contour line 74 of the individual image whose shape has already been specified. (Contour lines used in the individual image areas 53A to 53F) are also removed. Similarly, when the log shape is specified in the individual image areas 53B to 53F and 53H to 53L, the determined outline 74 is removed in addition to the outlines 71 to 73.
- the photometry / ranging unit 47 detects the distance (subject distance) to the log (object). Further, when the release button 18 is fully depressed to take a picture (S2), the image data of the picture taken by the CCD 31 is recorded in the buffer memory 42 and then taken into the individual image area extracting unit 51 (S3).
- the individual image region extraction unit 51 extracts the individual image regions 53A to 53L of logs in the captured image 52 by identifying the ellipse on the captured image 52 (S4).
- the individual line image generation unit 61 selects one of the individual image regions 53A to 53L extracted by the individual image region extraction unit 51 (for example, the individual image region 53A).
- the converted image extraction filter 63 numerically analyzes the color change in the individual image area 53A, and extracts an image having a color change larger than the extraction threshold.
- the individual line image generation unit 61 converts the extracted image into a line image and generates an individual line image 62 (S5).
- the removal unit 66 has a center at substantially the same position as the center of the individual line image 62 out of the contour lines included in the individual line image 62, and the vertical and horizontal lengths are approximately half of the individual line image 62.
- the following peripheral contour on the straight line connecting the center vicinity contour 71, the boundary contour 72 in contact with the boundary of the individual image area 53A, and the center point CP1 of the contour itself and the center point CP2 of the individual line image 62 The line 73 and the determined contour line 74 related to the estimated elliptical shape 76 whose shape has been specified before are removed (S6).
- the estimated shape determination unit 67 performs ellipse fitting on the corrected line image 64 from which each of the contour lines 71 to 74 has been removed, estimates the shape of the log, and determines the estimated ellipse shape 76 (S7).
- the estimated shape determination unit 81 detects the total number of pixels of the overlapping portion of the remaining outline 75 in the corrected line image 64 and the estimated elliptical shape 76, and determines whether the total number of pixels exceeds a threshold value. Determine (S8). When the total number of pixels exceeds the threshold (Y in S8), it is determined that the estimated ellipse shape 76 is appropriate, and the object shape specifying unit 83 specifies the estimated ellipse shape 76 as a log shape. (S9).
- the dimension measuring unit 84 measures the dimension of the identified estimated elliptical shape 76 (the dimension of the log on the captured image) (S10).
- the actual dimension calculation unit 85 calculates the actual dimension of the log based on the dimension measured by the dimension measurement unit 84, the distance to the log detected by the photometry / ranging unit 47, and the focal length (S11). ).
- the calculated actual size data of the log is recorded in the memory card 45.
- the outline near the center 71, the boundary outline 72, the peripheral outline 73, and the determined outline 74 are removed from the outlines included in the individual line image 62, and the outlines 71 to 74 are removed.
- the log shape is specified using the corrected line image 64. As a result, the shape of the log is specified more accurately than that in which the shape of the log is specified using the individual line image 62 including the contour lines 71 to 74.
- the corrected line image 64 is divided into four regions, and the estimated shape determining unit 67 selects at least one contour line from each region.
- symbol is attached
- the estimated shape determination unit 67 divides the corrected line image 64 into first to fourth areas 64a to 64d, and selects, for example, one contour line at a predetermined position (for example, the center) of each of the areas 64a to 64d. To do. Then, ellipse fitting is performed on the four selected contour lines to determine the estimated ellipse shape 76. As a result, the time taken to determine the estimated elliptical shape 76 is shorter than that in which ellipse fitting is performed on all contour lines in the corrected line image 64.
- the number of divisions of the corrected line image 64 is not limited to four divisions, and can be changed as appropriate. When dividing, it is preferable to divide evenly around the center point of the corrected line image 64. Further, the number of contour lines selected from each region is not limited to one and can be changed as appropriate.
- the digital camera 100 is provided with a distance calculation unit 101 that calculates a distance to a log (object).
- the user operates the menu key 21 to input the actual size of the log that is the object.
- the distance calculation unit 101 calculates the distance to the log based on the dimension measured by the dimension measurement unit 84, the actual log input size, and the focal length.
- the menu key 21 is operated to input the actual size of the target log (S102), and when the release button 18 is fully pressed (S103), the captured image data is extracted as an individual image area.
- the data is input to the unit 51 (S104).
- S105 to S111 and S113 are the same as S4 to S10 and S12 of the first embodiment, and thus description thereof is omitted.
- the distance calculating unit 101 calculates the dimension measured by the dimension measuring unit 84 and the input log. Based on the actual size and the focal length, the distance to the log is calculated (S112). The calculated distance data to the log is recorded in the memory card 45. S101 to S113 are performed for all of the individual image areas 53A to 53L.
- the digital camera 110 according to the fourth embodiment illustrated in FIGS. 12 and 13 generates a new individual line image 62 when the estimated shape determining unit 81 determines that the estimated elliptical shape 76 is not appropriate.
- the individual line image generation unit 111 has a plurality (for example, ten) of converted images for extracting images to be converted into line images from images in the selected individual image region. Extraction filters 112a to 112j are provided. The converted image extraction filters 112a to 112j have different extraction threshold values for extracting an image to be converted into a line image. The individual line image generation unit 111 extracts an image to be converted into a line image using any of the converted image extraction filters 112a to 112j. Then, the individual line image generation unit 111 generates the individual line image 62 by converting the extracted image into a line image.
- Extraction filters 112a to 112j are provided.
- the converted image extraction filters 112a to 112j have different extraction threshold values for extracting an image to be converted into a line image.
- the individual line image generation unit 111 extracts an image to be converted into a line image using any of the converted image extraction filters 112a to 112j. Then, the individual line image generation unit 111 generates the individual
- the individual line image generation unit 111 changes the one to be used among the converted image extraction filters 112a to 112j, and extracts the image again. A new individual line image 62 is generated. This is repeated until it is determined that the estimated elliptical shape 76 is appropriate.
- the individual line image generation unit 111 uses the converted image extraction filter 112a, the converted image extraction filter 112b, the converted image extraction filter 112c,..., The converted image extraction filter 112i, and the converted image extraction filter 112j in this order.
- the object shape specifying unit 83 specifies the estimated elliptical shape 76 having the highest evaluation result as a log shape. .
- the individual image area extraction unit 51 extracts log individual image areas 53A to 53L in the captured image 52 (S204).
- the individual line image generation unit 111 selects one of the individual image regions 53A to 53L (for example, the individual image region 53A) extracted by the individual image region extraction unit 51. Any of the converted image extraction filters 112a to 112j (for example, the converted image extraction filter 112a) numerically analyzes the color change in the individual image area 53A, and extracts an image having a color change larger than the extraction threshold value. .
- the individual line image generation unit 111 converts the extracted image into a line image and generates an individual line image 62 (S205). Since S206 to S211 are the same as S6 to S11 of the first embodiment, the description thereof is omitted.
- the conversion image extraction filters 112a to 112j to be used are changed (S212), and an image to be converted into a line image is extracted again.
- the individual line image generation unit 111 generates a new individual line image 62 based on the newly extracted image (S205). Then, S206 and subsequent steps are executed again. S201 to S212 are performed for all of the individual image areas 53A to 53L.
- the configuration of the fourth embodiment may be applied to the third embodiment to measure the distance to the log.
- the digital camera 120 of the fifth embodiment shown in FIGS. 14 and 15 generates a plurality of (for example, five) individual line images 62 from one individual image region.
- the individual line image generation unit 121 includes a plurality (for example, five) of converted image extraction filters 122a to 122e for extracting images to be converted into line images from images included in the selected individual image region. Is provided.
- the converted image extraction filters 122a to 122e have different extraction thresholds for extracting an image to be converted into a line image.
- the individual line image generation unit 121 extracts an image using each of the converted image extraction filters 122a to 122e, and generates five individual line images 62 corresponding to each.
- the estimated shape determining unit 67 determines five estimated elliptical shapes 76 using the five individual line images 62.
- the estimated shape determination unit 81 evaluates and determines these five estimated ellipse shapes 76.
- the object shape specifying unit 83 specifies the estimated elliptical shape 76 having the highest evaluation result in the estimated shape determining unit 81 as a log shape.
- the dimension measuring unit 84 measures the dimension of the specified estimated elliptical shape 76
- the captured image data is input to the individual image area extraction unit 51 (S303).
- the individual image area extraction unit 51 extracts log individual image areas 53A to 53L in the captured image 52 (S304).
- the individual line image generation unit 61 selects one of the individual image regions 53A to 53L extracted by the individual image region extraction unit 51 (for example, the individual image region 53A).
- Each of the converted image extraction filters 122a to 122e numerically analyzes the color change in the individual image area 53A, and extracts an image having a color change larger than the extraction threshold.
- the individual line image generation unit 111 converts the images extracted by the filters 122a to 122e into line images, respectively, and generates five individual line images 62 corresponding to the filters 122a to 122e (S305).
- the removal unit 66 removes the center vicinity contour line 71, the boundary contour line 72, the peripheral contour line 73, and the determined contour line 74 in each of the five individual line images 62 (S306).
- the estimated shape determination unit 67 performs ellipse fitting on each of the five corrected line images 64 to determine five estimated ellipse shapes 76 (S307).
- the estimated shape determination unit 81 detects the total number of pixels of the portion that overlaps the remaining outline 75 in each of the five estimated elliptic shapes 76. Then, based on the detected total number of pixels, the estimated shape determination unit 81 evaluates the five estimated ellipse shapes 76 (S308).
- the object shape specifying unit 83 specifies the estimated elliptical shape 76 having the highest evaluation result in the estimated shape determining unit 81 as a log shape (S309).
- the dimension measuring unit 84 measures the dimension of the specified estimated elliptical shape 76 (S310).
- the actual dimension calculation unit 85 calculates the actual dimension of the log based on the dimension measured by the dimension measurement unit 84, the distance to the log detected by the photometry / ranging unit 47, and the focal length (S311). ).
- S301 to S311 are performed for all of the individual image areas 53A to 53L.
- the configuration of the fifth embodiment may be applied to the third embodiment to measure the distance to the log.
- the digital camera 130 according to the sixth embodiment illustrated in FIGS. 16 and 17 has a threshold calculation unit 131 that calculates an extraction threshold for extracting an image to be converted into a line image from images included in the selected individual image region. Is provided.
- the threshold calculation unit 131 analyzes the hue in the selected individual image area, creates a histogram of the hue value, and adds a constant to the peak value of the histogram to set the extraction threshold. calculate.
- the extraction threshold may be calculated by subtracting a constant from the peak value. Further, a difference value or a ratio value having a constant frequency from the peak value, or a minimum difference value from the peak value may be used as the extraction threshold value.
- the individual line image generation unit 132 binarizes an image included in the individual image region (extracts an image having a higher hue value than the extraction threshold) based on the extraction threshold calculated by the threshold calculation unit 131.
- the individual line image 62 is generated.
- the threshold value calculation unit 131 changes the constant added to the peak value and calculates the extraction threshold value again. Then, the individual line image generation unit 132 generates a new individual line image 62 based on the new extraction threshold. This is repeated until it is determined that the estimated elliptical shape 76 is appropriate.
- the release button 18 is fully pressed in the log measurement mode (S401) (S402)
- the captured image data is input to the individual image area extraction unit 51 (S403).
- the individual image area extraction unit 51 extracts log individual image areas 53A to 53L in the captured image 52 (S404).
- the individual line image generation unit 132 selects one of the individual image regions 53A to 53L extracted by the individual image region extraction unit 51 (for example, the individual image region 53A).
- the threshold calculation unit 131 analyzes the hue in the individual image region 53A and creates a histogram of the hue value.
- the threshold calculation unit 131 calculates an extraction threshold by adding a constant to the peak value of the created histogram (S405).
- the individual line image generation unit 132 extracts an image having a hue value higher than the extraction threshold value calculated by the threshold value calculation unit 131 from the images included in the individual image region 53A.
- the individual line image generation unit 132 converts the extracted image into a line image and generates the individual line image 62 (S406). Note that S407 to S412 are the same as S6 to S11 of the first embodiment, and thus the description thereof is omitted.
- the threshold value calculation unit 131 changes a constant added to the histogram peak value of the hue value of the individual image area 53A, and calculates a new extraction threshold value (S413).
- the individual line image generation unit 132 extracts an image from the individual image region 53A using the new extraction threshold value, and generates a new individual line image 62 based on the newly extracted image (S406). Then, S407 and subsequent steps are executed again. S401 to S413 are performed for all of the individual image areas 53A to 53L.
- the configuration of the sixth embodiment may be applied to the third embodiment to measure the distance to the log.
- the digital camera has been described.
- the present invention can also be applied to an image processing apparatus (system) including a digital camera and a personal computer.
- the personal computer extracts the individual image area, generates the individual line image, removes the unnecessary contour line, determines the estimated shape, determines the suitability of the estimated shape, and the target object.
- the shape is specified and the dimensions of the object are measured.
- the present invention can be applied to a stereo camera that captures images from a plurality of viewpoints and acquires a plurality of viewpoint images, a mobile phone with a camera, and the like.
- a stereo camera When applied to a stereo camera, a normalized image that is an image viewed from the front is generated based on a plurality of viewpoint images, and each process is performed based on the normalized image.
- the shape of the object is specified, and the actual size of the object or the distance to the object is measured based on the shape.
- the center vicinity contour line, the boundary contour line, the peripheral contour line, and the determined contour line are removed from the contour lines included in the individual line image. What is necessary is just to remove at least one of the lines.
- the selection of each individual image area may be performed, for example, by measuring the size of logs in the order of the degree of coincidence of the shapes or in the order of the resolution. Furthermore, the user may determine the order.
- the suitability of the estimated ellipse shape is determined according to the total number of pixels in the portion where the estimated ellipse shape overlaps the contour line in the corrected line image.
- the suitability determination may be performed according to the number of pixels of the contour line within a predetermined distance from the estimated ellipse shape. In this case, it is preferable to weight the contour line in the order from the estimated ellipse shape.
- the threshold calculation unit calculates the extraction threshold again, but the threshold calculation unit calculates a plurality of extraction thresholds in advance.
- a plurality of individual line images may be generated using the plurality of extraction threshold values.
- the actual size of the log is calculated using the individual line image having the highest evaluation result among the plurality of individual line images.
- the threshold value calculation unit calculates the extraction threshold value.
- a plurality of extraction threshold values may be set in advance.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Botany (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
Description
図1に示すように、デジタルカメラ10は、カメラ本体10aの前面に、レンズ鏡筒11が設けられている。レンズ鏡筒11には、撮像レンズ12が組み込まれている。レンズ鏡筒11は、デジタルカメラ10が電源オフ状態にある際に、カメラ本体10aの内部に収納され、電源オン状態ではカメラ本体10aの前面から突出してワイド位置にセットされる。撮像レンズ12として、変倍レンズ12a、フォーカスレンズ12b(図3参照)を有するズームレンズが用いられる。
図9に示す第2実施形態は、修正済み線画像64を4つの領域に分割し、推定形状決定部67は各領域から少なくとも1つずつ輪郭線を選択する。なお、第1実施形態と同じものには同じ符号を付し、その説明を省略する。
図10及び図11に示す第3実施形態のデジタルカメラ100は、測光・測距部が省略され、寸法測定部84で測定された寸法を用いて、丸太までの距離を算出する。なお、第1実施形態と同じものには同じ符号を付し、その説明を省略する。
図12及び図13に示す第4実施形態のデジタルカメラ110は、推定形状判定部81で推定楕円形状76が適正でないと判定されたときに、新たな個別線画像62を生成する。
図14及び図15に示す第5実施形態のデジタルカメラ120は、1個の個別画像領域から複数(例えば、5個)の個別線画像62を生成する。
図16及び図17に示す第6実施形態のデジタルカメラ130は、選択した個別画像領域に含まれる画像の中から線画像に変換する画像を抽出するための抽出用閾値を算出する閾値算出部131を備える。
54A~54L 丸太画像
64a~64d 第1~第4領域
71 中心付近輪郭線
72 境界輪郭線
73 周辺輪郭線
74 決定済み輪郭線
75 残存輪郭線
76 推定楕円形状
Claims (16)
- 特定形状を有する対象物と推定される複数の個別画像を画面内から識別し、前記各個別画像が存在する個別画像領域を前記画面内から抽出する個別画像領域抽出部と、
前記各個別画像領域を順番に選択し、選択された前記個別画像領域内の前記個別画像を含む画像をそれぞれ線画像に変換して、個別線画像を生成する個別線画像生成部と、
前記各個別線画像に含まれる輪郭線の中から、前記個別線画像の中心とほぼ同じ位置に中心を有し、上下方向及び左右方向の長さが前記個別線画像のほぼ半分以下の輪郭線、及び/又は前記輪郭線自身の中心点と前記個別線画像の中心点とを結ぶ直線上に存在する輪郭線を、不要な輪郭線として除去する除去部と、
前記不要な輪郭線を除去した修正済み線画像に対応するように、前記特定形状を最適な大きさとした推定形状を定める推定形状決定部と、
前記推定形状と前記修正済み線画像との重なり度合いを評価して、前記推定形状の適否を判定する判定部と、
前記推定形状が適正であると判定されたときに、前記推定形状を前記画面から抽出された前記対象物の形状として特定する対象物形状特定部と、
を備えることを特徴とする形状抽出装置。 - 前記除去部は、前記個別画像領域が重複している場合に、先に形状が特定された前記対象物に係る前記修正済み線画像の輪郭線も除去することを特徴とする請求の範囲第1項記載の形状抽出装置。
- 前記除去部は、前記個別画像領域の境界に接している輪郭線も除去することを特徴とする請求の範囲第1項または第2項記載の形状抽出装置。
- 前記推定形状決定部は、前記修正済み線画像を複数の領域に分けて、複数の領域それぞれから輪郭線を選択し、この選択した複数の輪郭線に基づいて前記推定形状を決定することを特徴とする請求の範囲第1項ないし第3項いずれか1つ記載の形状抽出装置。
- 前記推定形状が適正でないと判定されたときに、前記除去部は、前記推定形状の外側の輪郭線を除去することを特徴とする請求の範囲第1項ないし第4項いずれか1つ記載の形状抽出装置。
- 前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像の中から線画像に変換する画像を抽出するための閾値がそれぞれ異なる複数の変換画像抽出フィルタを有し、前記複数の変換画像抽出フィルタのいずれかを用いて抽出された画像を線画像に変換して前記個別線画像を生成し、
前記推定形状が適正でないと判定されたときに、前記個別線画像生成部は、前記変換画像抽出フィルタを変更して新たな個別線画像を生成することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像の中から線画像に変換する画像を抽出するための閾値がそれぞれ異なる複数の変換画像抽出フィルタを有し、前記各変換画像抽出フィルタを用いて抽出された画像をそれぞれ線画像に変換して、1つの前記個別画像領域に対して複数の前記個別線画像を生成し、
前記推定形状決定部は、前記複数の個別線画像それぞれを用いて複数の推定形状を決定し、
前記判定部は、前記複数の推定形状それぞれを判定し、
前記対象物形状特定部は、前記複数の推定形状のうち、前記判定部での評価が最も高い1つを前記対象物の形状として特定することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像を2値化して前記個別線画像を生成するための閾値を複数有し、前記選択された個別画像領域に含まれる画像を前記複数の閾値のいずれかに基づいて2値化して前記個別線画像を生成し、
前記推定形状が適正でないと判定されたときに、前記個別線画像生成部は、前記閾値を変更して新たな個別線画像を生成することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像を2値化して前記個別線画像を生成するための閾値を複数有し、前記選択された個別画像領域に含まれる画像を前記複数の閾値それぞれに基づいて2値化して、1つの前記個別画像領域に対して複数の前記個別線画像を生成し、
前記推定形状決定部は、前記複数の個別線画像それぞれを用いて複数の推定形状を決定し、
前記判定部は、前記複数の推定形状それぞれを判定し、
前記対象物形状特定部は、前記複数の推定形状のうち、前記判定部での評価が最も高い1つを前記対象物の形状として特定することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 前記選択された個別画像領域に含まれる画像を2値化して前記個別線画像を生成するための閾値を算出する閾値算出部を備え、
前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像を、前記閾値算出部で算出された閾値に基づいて2値化して前記個別線画像を生成し、
前記推定形状が適正でないと判定されたときに、前記閾値算出部は、新たな閾値を算出し、前記個別線画像生成部は、前記新たな閾値に基づいて新たな個別線画像を生成することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 前記選択された個別画像領域に含まれる画像を2値化して前記個別線画像を生成するための閾値を複数算出する閾値算出部を備え、
前記個別線画像生成部は、前記選択された個別画像領域に含まれる画像を前記複数の閾値それぞれに基づいて2値化して、1つの前記個別画像領域に対して複数の前記個別線画像を生成し、
前記推定形状決定部は、前記複数の個別線画像それぞれを用いて複数の推定形状を決定し、
前記判定部は、前記複数の推定形状それぞれを判定し、
前記対象物形状特定部は、前記複数の推定形状のうち、前記判定部での評価が最も高い1つを前記対象物の形状として特定することを特徴とする請求の範囲第1項ないし第5項いずれか1つ記載の形状抽出装置。 - 特定形状を有する対象物と推定される個別画像を画面内から識別し、前記各個別画像が存在する個別画像領域を前記画面内から抽出する個別画像領域抽出部と、
前記各個別画像領域を順番に選択し、選択された前記個別画像領域内の前記個別画像を含む画像をそれぞれ線画像に変換して、個別線画像を生成する個別線画像生成部と、
前記各個別線画像に含まれる輪郭線の中から、前記個別線画像の中心とほぼ同じ位置に中心を有し、上下方向及び左右方向の長さが前記個別線画像のほぼ半分以下の輪郭線、及び/又は前記輪郭線自身の中心点と前記個別線画像の中心点とを結ぶ直線上に存在する輪郭線を、不要な輪郭線として除去する除去部と、
前記不要な輪郭線を除去した修正済み線画像に対応するように、前記特定形状を最適な大きさとした推定形状を定める推定形状決定部と、
前記推定形状と前記修正済み線画像との重なり度合いを評価して、前記推定形状の適否を判定する判定部と、
前記推定形状が適正であると判定されたときに、前記推定形状を前記画面から抽出された前記対象物の形状として特定する対象物形状特定部と、
前記対象物を撮影するときの前記対象物までの距離を測定する距離測定部と、
前記対象物形状特定部で特定された対象物形状の寸法を測定する寸法測定部と、
前記距離測定部で測定された距離と、前記寸法測定部で測定された寸法とに基づいて、前記対象物の実寸法を算出する実寸法算出部と、
を備えることを特徴とする寸法測定装置。 - 前記対象物を撮影するときの撮影倍率を入力する倍率入力部を備え、
前記実寸法算出部は、前記距離測定部で測定された距離と、前記寸法測定部で測定された寸法と、前記倍率入力部から入力された撮影倍率とに基づいて、前記対象物の実寸法を算出することを特徴とする請求の範囲第12項記載の寸法測定装置。 - 特定形状を有する対象物と推定される個別画像を画面内から識別し、前記各個別画像が存在する個別画像領域を前記画面内から抽出する個別画像領域抽出部と、
前記各個別画像領域を順番に選択し、選択された前記個別画像領域内の前記個別画像を含む画像をそれぞれ線画像に変換して、個別線画像を生成する個別線画像生成部と、
前記各個別線画像に含まれる輪郭線の中から、前記個別線画像の中心とほぼ同じ位置に中心を有し、上下方向及び左右方向の長さが前記個別線画像のほぼ半分以下の輪郭線、及び/又は前記輪郭線自身の中心点と前記個別線画像の中心点とを結ぶ直線上に存在する輪郭線を、不要な輪郭線として除去する除去部と、
前記不要な輪郭線を除去した修正済み線画像に対応するように、前記特定形状を最適な大きさとした推定形状を定める推定形状決定部と、
前記推定形状と前記修正済み線画像との重なり度合いを評価して、前記推定形状の適否を判定する判定部と、
前記推定形状が適正であると判定されたときに、前記推定形状を前記画面から抽出された前記対象物の形状として特定する対象物形状特定部と、
前記対象物の実寸法を入力する実寸法入力部と、
前記対象物形状特定部で特定された対象物形状の寸法を測定する寸法測定部と、
前記実寸法入力部で入力された実寸法と、前記寸法測定部で測定された寸法とに基づいて、前記対象物を撮影するときの前記対象物までの距離を算出する距離算出部と、
を備えることを特徴とする距離測定装置。 - 前記対象物を撮影するときの撮影倍率を入力する倍率入力部を備え、
前記距離算出部は、前記実寸法入力部で入力された実寸法と、前記寸法測定部で測定された寸法と、前記倍率入力部から入力された撮影倍率とに基づいて、前記対象物までの距離を算出することを特徴とする請求の範囲第14項記載の距離測定装置。 - 特定形状を有する対象物と推定される個別画像を画面内から識別し、前記各個別画像が存在する個別画像領域を前記画面内から抽出する個別画像領域抽出ステップと、
前記各個別画像領域を順番に選択し、選択された前記個別画像領域内の前記個別画像を含む画像をそれぞれ線画像に変換して、個別線画像を生成する個別線画像生成ステップと、
前記各個別線画像に含まれる輪郭線の中から、前記個別線画像の中心とほぼ同じ位置に中心を有し、上下方向及び左右方向の長さが前記個別線画像のほぼ半分以下の輪郭線、及び/又は前記輪郭線自身の中心点と前記個別線画像の中心点とを結ぶ直線上に存在する輪郭線を、不要な輪郭線として除去する除去ステップと、
前記不要な輪郭線を除去した修正済み線画像に対応するように、前記特定形状を最適な大きさとした推定形状を定める推定形状決定ステップと、
前記推定形状と前記修正済み線画像との重なり度合いを評価して、前記推定形状の適否を判定する判定ステップと、
前記推定形状が適正であると判定されたときに、前記推定形状を前記画面から抽出された前記対象物の形状として特定する対象物形状特定ステップと、
を備えることを特徴とする形状抽出方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/641,452 US8855417B2 (en) | 2010-06-29 | 2011-05-24 | Method and device for shape extraction, and size measuring device and distance measuring device |
CN201180022050.4A CN102870135B (zh) | 2010-06-29 | 2011-05-24 | 用于形状提取的方法和装置、尺寸测量装置和距离测量装置 |
JP2012522514A JP5620990B2 (ja) | 2010-06-29 | 2011-05-24 | 形状抽出方法及び装置、並びに寸法測定装置及び距離測定装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2010-147804 | 2010-06-29 | ||
JP2010147804 | 2010-06-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012002069A1 true WO2012002069A1 (ja) | 2012-01-05 |
Family
ID=45401803
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/061800 WO2012002069A1 (ja) | 2010-06-29 | 2011-05-24 | 形状抽出方法及び装置、並びに寸法測定装置及び距離測定装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8855417B2 (ja) |
JP (1) | JP5620990B2 (ja) |
CN (1) | CN102870135B (ja) |
WO (1) | WO2012002069A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015191572A (ja) * | 2014-03-28 | 2015-11-02 | 株式会社パスコ | マンホール検出装置、マンホール検出方法及びプログラム |
WO2016194576A1 (ja) * | 2015-05-29 | 2016-12-08 | ソニー株式会社 | 情報処理装置および方法 |
JP2018076333A (ja) * | 2012-02-09 | 2018-05-17 | ヴァー2・ファーマシューティカルズ・アンパルトセルスカブVAR2 Pharmaceuticals ApS | コンドロイチン硫酸グリカンのターゲティング |
CN111028239A (zh) * | 2019-08-10 | 2020-04-17 | 杭州屏行视界信息科技有限公司 | 一种面向特制量体衣的椭圆精确识别方法 |
JP2021177726A (ja) * | 2020-05-14 | 2021-11-18 | 株式会社藤興業 | 木材切断端面直径の撮像計測方法及び装置 |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9355495B2 (en) * | 2013-10-09 | 2016-05-31 | Trimble Navigation Limited | Method and system for 3D modeling using feature detection |
CN106373156A (zh) * | 2015-07-20 | 2017-02-01 | 小米科技有限责任公司 | 通过图像确定空间参数的方法、装置及终端设备 |
CN106292159A (zh) * | 2016-09-26 | 2017-01-04 | 北京小米移动软件有限公司 | 微型投影仪 |
CN111512115B (zh) | 2018-09-17 | 2021-11-05 | 数据储存有限责任公司 | 原木检尺系统及相关方法 |
CN109645648B (zh) * | 2018-12-08 | 2022-02-11 | 余姚市恒正金属制品有限公司 | 扩展型儿童式冲头床 |
CN109631766B (zh) * | 2018-12-11 | 2021-01-19 | 南京第五十五所技术开发有限公司 | 一种基于图像的木材板尺寸测量方法 |
CN112037195B (zh) * | 2020-08-31 | 2023-04-07 | 中冶赛迪信息技术(重庆)有限公司 | 一种棒材长度异常的检测方法、系统、设备及介质 |
CN113781481A (zh) * | 2021-11-11 | 2021-12-10 | 滨州学院 | 非接触测量物体形状和尺寸的方法、装置以及电子设备 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0655142A (ja) * | 1992-08-06 | 1994-03-01 | Nanbu:Kk | 木材選別装置 |
JP2001314374A (ja) * | 2000-05-09 | 2001-11-13 | Topcon Corp | 角膜内皮細胞測定装置 |
JP2003031469A (ja) * | 2001-07-13 | 2003-01-31 | Toshiba Corp | パターン評価方法,位置合わせ方法、検査装置の検査方法,半導体製造工程の管理方法 |
JP2003247954A (ja) * | 2002-02-26 | 2003-09-05 | Sumitomo Metal Mining Co Ltd | 円形体周縁部の欠陥検出方法 |
JP2007246031A (ja) * | 2006-03-17 | 2007-09-27 | Matsushita Electric Ind Co Ltd | 車両管理装置 |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5619593A (en) * | 1991-09-12 | 1997-04-08 | Fuji Photo Film Co., Ltd. | Method for extracting object images and method for detecting movements thereof |
JP3679512B2 (ja) * | 1996-07-05 | 2005-08-03 | キヤノン株式会社 | 画像抽出装置および方法 |
JP4143149B2 (ja) | 1997-11-20 | 2008-09-03 | キヤノン株式会社 | 領域抽出装置、領域抽出方法及びコンピュータ読み取り可能な記録媒体 |
US6778690B1 (en) * | 1999-08-13 | 2004-08-17 | Hanif M. Ladak | Prostate boundary segmentation from 2D and 3D ultrasound images |
WO2003070102A2 (en) * | 2002-02-15 | 2003-08-28 | The Regents Of The University Of Michigan | Lung nodule detection and classification |
GB0305315D0 (en) * | 2003-03-07 | 2003-04-09 | Weber Martin | Image processing system |
US7558402B2 (en) * | 2003-03-07 | 2009-07-07 | Siemens Medical Solutions Usa, Inc. | System and method for tracking a global shape of an object in motion |
US7545965B2 (en) * | 2003-11-10 | 2009-06-09 | The University Of Chicago | Image modification and detection using massive training artificial neural networks (MTANN) |
US8073226B2 (en) * | 2006-06-30 | 2011-12-06 | University Of Louisville Research Foundation, Inc. | Automatic detection and monitoring of nodules and shaped targets in image data |
JP2008059081A (ja) * | 2006-08-29 | 2008-03-13 | Sony Corp | 画像処理装置及び画像処理方法、並びにコンピュータ・プログラム |
JP5197205B2 (ja) * | 2008-07-25 | 2013-05-15 | 株式会社島精機製作所 | 輪郭抽出装置と輪郭抽出方法及び輪郭抽出プログラム |
JP5052451B2 (ja) * | 2008-07-30 | 2012-10-17 | オリンパス株式会社 | 細胞測定装置および細胞測定方法 |
US8175379B2 (en) * | 2008-08-22 | 2012-05-08 | Adobe Systems Incorporated | Automatic video image segmentation |
CN102113897B (zh) * | 2009-12-31 | 2014-10-15 | 深圳迈瑞生物医疗电子股份有限公司 | 一种在图像中提取及测量感兴趣目标的方法及其装置 |
-
2011
- 2011-05-24 US US13/641,452 patent/US8855417B2/en not_active Expired - Fee Related
- 2011-05-24 CN CN201180022050.4A patent/CN102870135B/zh not_active Expired - Fee Related
- 2011-05-24 JP JP2012522514A patent/JP5620990B2/ja not_active Expired - Fee Related
- 2011-05-24 WO PCT/JP2011/061800 patent/WO2012002069A1/ja active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0655142A (ja) * | 1992-08-06 | 1994-03-01 | Nanbu:Kk | 木材選別装置 |
JP2001314374A (ja) * | 2000-05-09 | 2001-11-13 | Topcon Corp | 角膜内皮細胞測定装置 |
JP2003031469A (ja) * | 2001-07-13 | 2003-01-31 | Toshiba Corp | パターン評価方法,位置合わせ方法、検査装置の検査方法,半導体製造工程の管理方法 |
JP2003247954A (ja) * | 2002-02-26 | 2003-09-05 | Sumitomo Metal Mining Co Ltd | 円形体周縁部の欠陥検出方法 |
JP2007246031A (ja) * | 2006-03-17 | 2007-09-27 | Matsushita Electric Ind Co Ltd | 車両管理装置 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018076333A (ja) * | 2012-02-09 | 2018-05-17 | ヴァー2・ファーマシューティカルズ・アンパルトセルスカブVAR2 Pharmaceuticals ApS | コンドロイチン硫酸グリカンのターゲティング |
JP2015191572A (ja) * | 2014-03-28 | 2015-11-02 | 株式会社パスコ | マンホール検出装置、マンホール検出方法及びプログラム |
WO2016194576A1 (ja) * | 2015-05-29 | 2016-12-08 | ソニー株式会社 | 情報処理装置および方法 |
CN111028239A (zh) * | 2019-08-10 | 2020-04-17 | 杭州屏行视界信息科技有限公司 | 一种面向特制量体衣的椭圆精确识别方法 |
JP2021177726A (ja) * | 2020-05-14 | 2021-11-18 | 株式会社藤興業 | 木材切断端面直径の撮像計測方法及び装置 |
Also Published As
Publication number | Publication date |
---|---|
JP5620990B2 (ja) | 2014-11-05 |
JPWO2012002069A1 (ja) | 2013-08-22 |
CN102870135B (zh) | 2016-02-03 |
US20130034270A1 (en) | 2013-02-07 |
US8855417B2 (en) | 2014-10-07 |
CN102870135A (zh) | 2013-01-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5620990B2 (ja) | 形状抽出方法及び装置、並びに寸法測定装置及び距離測定装置 | |
JP5762211B2 (ja) | 画像処理装置および画像処理方法、プログラム | |
KR101396743B1 (ko) | 화상 처리 장치, 화상 처리 방법, 및 기록 매체 | |
US9451150B2 (en) | Image capturing apparatus comprising focus detection, and method for controlling the same | |
JP5882975B2 (ja) | 画像処理装置、撮像装置、画像処理方法、及び記録媒体 | |
CN104519328B (zh) | 图像处理设备、图像捕捉装置和图像处理方法 | |
CN103024265B (zh) | 摄像装置和摄像装置的摄像方法 | |
JP5843454B2 (ja) | 画像処理装置、画像処理方法およびプログラム | |
CN102959942B (zh) | 立体观看用图像拍摄装置及其控制方法 | |
JP7223079B2 (ja) | 画像処理装置およびその制御方法、ならびに撮像装置 | |
JP4042750B2 (ja) | 画像処理装置、コンピュータプログラム、及び画像処理方法 | |
CN111385471B (zh) | 信息处理设备和方法、摄像设备及其控制方法、存储介质 | |
US8295609B2 (en) | Image processing apparatus, image processing method and computer readable-medium | |
CN102542251B (zh) | 被摄体检测装置以及被摄体检测方法 | |
JP5619124B2 (ja) | 画像処理装置、撮像装置、画像処理プログラムおよび画像処理方法 | |
JP2013542528A (ja) | 夜景画像ボケ検出システム | |
JP2001249265A (ja) | 測距装置 | |
JP2013187645A (ja) | ノイズ低減装置およびノイズ低減方法 | |
JP4771536B2 (ja) | 撮像装置及び主被写体となる顔の選択方法 | |
JP5012477B2 (ja) | 部分画像抽出方法および部分画像抽出装置 | |
US8675958B2 (en) | Subject determination method, computer program product for determining subject, and camera | |
JP5157673B2 (ja) | 撮像装置、及びフォーカス位置表示方法、フォーカス位置表示プログラム | |
JP2013005405A (ja) | 電子カメラ | |
JP6421032B2 (ja) | 焦点検出装置、焦点検出方法及び焦点検出プログラム | |
JP5928024B2 (ja) | 画像処理装置、画像処理方法及びプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201180022050.4 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11800538 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012522514 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13641452 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11800538 Country of ref document: EP Kind code of ref document: A1 |