WO2018068414A1 - 色环电阻的检测方法、装置和自动光学检测系统 - Google Patents
色环电阻的检测方法、装置和自动光学检测系统 Download PDFInfo
- Publication number
- WO2018068414A1 WO2018068414A1 PCT/CN2016/113145 CN2016113145W WO2018068414A1 WO 2018068414 A1 WO2018068414 A1 WO 2018068414A1 CN 2016113145 W CN2016113145 W CN 2016113145W WO 2018068414 A1 WO2018068414 A1 WO 2018068414A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- color
- color ring
- image
- ring resistance
- value
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Definitions
- the present invention relates to the field of electronic technology, and in particular, to a method for detecting a color ring resistance, a color ring resistance detecting device, and an automatic optical detecting system.
- AOI Automated Optical Inspection
- Error, leakage and back detection of electronic components is a common application in the field of board defect detection.
- the error, leakage, and back detection of color ring resistance are extremely important.
- the defect detection of the color ring resistance of the AOI system is generally concentrated on the detection of the missing parts, and the detection of the wrong part and the reverse part of the color ring resistance needs to be manually performed by means of an additional color ring resistance identification auxiliary device, such as a color ring. Resistor direct reading card, color ring resistance identification instrument, etc., this solution requires extra hardware and manpower, and the cost is high.
- a method for detecting color ring resistance comprising the steps of:
- the color ring resistance is detected according to a color value of the color ring.
- the above method for detecting the color ring resistance automatically scanning the circuit board to acquire an image, and then extracting an image of the color ring resistance, and Through the image processing technology, the color value of the color ring included in the color ring resistance is further extracted, thereby determining whether the color ring resistance has faults and reverse defects according to the color values of the respective color rings, thereby realizing the automatic detection of the color ring resistance, which is no longer needed.
- additional color ring resistance to identify the auxiliary equipment, there is no need to arrange additional workers for inspection, which effectively saves hardware and labor costs.
- the speed of the color ring resistance detection is no longer limited by the skill of the worker, and the detection speed of the color ring resistance is greatly improved.
- the determining, according to the image of the color ring resistance, the candidate region of the color ring included in the color ring resistance comprises: determining, according to the image of the color ring resistance, the color ring included in the color ring resistance An abscissa; extending a preset length in a horizontal direction with the abscissa of the color circle as a center point, obtaining a width of the color ring; obtaining a width according to the width of the color ring and the image of the color ring resistance
- the color ring resistance includes a candidate region of a color circle. According to the image of the extracted color ring resistance, the area of the color ring included in the color ring resistance is automatically located, and the accuracy is high.
- the method before the preset length is extended in the horizontal direction with the abscissa of the color circle as a center point, the method further includes the steps of: according to the preset color circle width and the preset adjacent two color rings The distance determines the preset length.
- the length of the color circle needs to be extended according to the preset color circle width and the preset distance between two adjacent color rings, so that the width of the color ring can be accurately obtained.
- the step of determining an abscissa of the color ring included in the color ring resistance according to the image of the color ring resistance comprises: converting an image of the color ring resistance into an RGB color mode corresponding to each color channel a sub-image; performing color circle edge detection in the X direction for each sub image, synthesizing the first edge image according to the color ring edge detection result of all the sub images; binarizing the first edge image to obtain the second edge image And summing the values of the points in the second edge image in the Y direction to obtain a first vector; each of the first vectors is a value summed in the Y direction corresponding to each abscissa; The peak position in the second edge image is determined by the first vector, and the X coordinate of the peak position is determined as the abscissa of the color ring included in the color ring resistance.
- the step of synthesizing the first edge image according to the color circle edge detection result of all the sub images includes: obtaining a maximum value of the same color ring in the color circle edge detection result of all the sub images; according to the obtained color circles The corresponding maximum value synthesizes the first edge image.
- the image after the edge detection of the sub-channel is fused according to the maximum value, and then the abscissa of the color circle is determined according to the fused image, thereby effectively improving the accuracy of determining the horizontal coordinate of the color circle.
- the method includes the step of setting a value of a point in the first edge image after binarization that is smaller than a set distance between two edges in the horizontal direction to be 0. Considering that the color ring does not appear at the left and right edges of the color ring resistance, the value near the left and right edges of the binarized image is set to 0, thereby effectively filtering the interference information, thereby effectively improving the accuracy of determining the horizontal axis of the color circle.
- the method further includes the step of: setting a value of the first vector that is less than a preset threshold to 0. A value smaller than the preset threshold in the first vector is set to 0, so that the peak position corresponding to the excellent ring can be more accurately located.
- the color region of the candidate region of the color circle is color-recognized
- the step of obtaining a color value of the color circle includes: averaging values of points in the candidate region of the color circle in the Y direction, Obtaining a second vector; each of the second vectors is a value averaged in a Y direction corresponding to each abscissa; and comparing each value in the second vector with a preset color value Determining a preset color value when the difference is the smallest as a color value of the corresponding value in the second vector; and filtering out a color of the second color that has the most occurrences and is not the color of the body of the color ring resistance A value, the filtered color value is used as the color value of the color circle, and the body color is the color of the color ring resistance surface other than the color of the color ring.
- the color circle candidate region may also have a body color of the colored ring resistance, and the color of each pixel point on the color ring may also be somewhat deviated due to noise
- the color value of the candidate region color color obtained in the ordinate direction is The averaging is performed to determine the color of the color circle according to the average value of the color, and the accuracy of determining the color value of the color circle is improved.
- the step of extracting an image of the color ring resistance from the image of the circuit board includes: obtaining the circuit according to a position of the color ring resistance set in the circuit board when the circuit board is plated a preliminary position of the color ring resistance in the image of the board; expanding the preliminary position of the color ring resistance outward to each of the preset pixels to obtain a candidate area of the color ring resistance; according to the color when the circuit board is used for plate making An image of the ring resistance is image-matched to the candidate region of the color ring resistance to obtain an image of the color ring resistance in the candidate region of the color ring resistance.
- the position of the color ring resistance in the image of the production circuit board is expanded to obtain a candidate region of the color ring resistance, and an image of the color ring resistance is extracted according to the candidate region. , greatly improving the efficiency and accuracy of color ring resistance image extraction.
- a color ring resistance detecting device comprising:
- a color ring resistance image extraction module configured to acquire an image of the circuit board to be detected, and extract an image of the color ring resistance from the image of the circuit board;
- a color circle candidate region determining module configured to determine a candidate region of the color ring included in the color ring resistance according to the image of the color ring resistance;
- a color ring color value obtaining module configured to perform color recognition on a candidate area of the color ring to obtain a color value of the color ring;
- the color ring resistance detecting module is configured to detect the color ring resistance according to the color value of the color ring.
- the above-mentioned color ring resistance detecting device automatically scans the circuit board to acquire an image, and then extracts an image of the color ring resistance, and further extracts the color value of the color ring included in the color ring resistance by image processing technology, thereby according to the color value of each color ring It is judged whether the color ring resistance has faults and reverse defects, and the automatic detection of the color ring resistance is realized. It is no longer necessary to identify the auxiliary equipment by means of additional color ring resistance, and there is no need to arrange additional workers for detection, thereby effectively saving hardware cost. And labor costs. In addition, since the color ring resistance is no longer required to be detected manually, the speed of the color ring resistance detection is no longer limited by the skill of the worker, and the detection speed of the color ring resistance is greatly improved.
- the color circle candidate region determining module includes: a color circle abscissa determining unit, configured to determine an abscissa of a color ring included in the color ring resistance according to an image of the color ring resistance; a color circle width a determining unit, configured to extend a preset length in the horizontal direction with the abscissa of the color circle as a center point, to obtain a width of the color ring; a color circle candidate region determining unit, configured to use the width of the color ring and The height of the image of the color ring resistance obtains a candidate region of the color ring included in the color ring resistance.
- the color circle candidate region determining module automatically locates the region of the color ring included in the color ring resistance according to the image of the extracted color ring resistance, and the accuracy is high.
- the color circle abscissa determining unit includes: a sub-image obtaining sub-unit, configured to convert an image of the color ring resistance into a sub-image corresponding to each color channel according to an RGB color mode; the first edge image Obtaining a sub-unit for respectively performing color circle edge detection in the X direction for each sub image, synthesizing the first edge image according to the color circle edge detection result of all the sub images; and obtaining a subunit for the second edge image An edge image is binarized to obtain a second edge image; the first vector obtaining subunit is configured to sum the values of the points in the second edge image in the Y direction to obtain a first vector; Each value in a vector is a value summed in the Y direction corresponding to each abscissa; an abscissa determining subunit is configured to determine a peak position in the second edge image from the first vector, and to peak The X coordinate of the position is determined as the abscissa of the color ring included in
- the color ring color value obtaining module includes: a second vector obtaining unit for the color The values of the points in the candidate region of the ring are averaged in the Y direction to obtain a second vector; each of the second vectors is a value averaged in the Y direction corresponding to each abscissa; the color value comparison unit And comparing each value in the second vector with a preset color value, and determining a preset color value when the difference is minimum as a color value of a corresponding value in the second vector; a ring color value determining unit, configured to filter out a color value of a color of a subject that has the most occurrences in the second vector and is not the color ring resistance, and uses the filtered color value as a color value of the color ring.
- the body color is the color of the color ring resistance surface other than the color of the color ring.
- the color circle candidate region may also have a body color of the colored ring resistance, and the color of each pixel point on the color ring may also be somewhat deviated due to noise, the color ring color value obtaining module candidates for the obtained color circle are obtained.
- the area color value is averaged in the ordinate direction, thereby determining the color of the color ring according to the average value of the color, and improving the accuracy of determining the color value of the color circle.
- An automatic optical detection system comprising the color ring resistance detecting device according to any one of the above.
- the automatic optical detection system can automatically scan the circuit board to acquire an image, and then extract an image of the color ring resistance, and further extract the color value of the color ring included in the color ring resistance through image processing technology, thereby determining the color according to the color value of each color ring. Whether the ring resistor has faults and reverse defects, and the automatic detection of the color ring resistance is realized. It is no longer necessary to identify the auxiliary equipment by means of additional color ring resistance, and there is no need to arrange additional workers for inspection, which effectively saves hardware cost and labor. cost. In addition, since the color ring resistance is no longer required to be detected manually, the speed of the color ring resistance detection is no longer limited by the skill of the worker, and the detection speed of the color ring resistance is greatly improved.
- FIG. 1 is a schematic flow chart of a method for detecting color ring resistance in an embodiment
- FIG. 2 is a schematic flow chart of a method for extracting a color ring resistance image in an embodiment
- FIG. 3 is a schematic flow chart of a method for determining a color circle candidate region in an embodiment
- FIG. 4 is a schematic flow chart of a method for determining an abscissa of a color circle in an embodiment
- Figure 5 is a schematic illustration of three sub-images divided according to RGB in a particular embodiment
- FIG. 6 is a schematic diagram of three edge sub-images obtained by performing edge detection of three sub-images in a horizontal direction in a specific embodiment
- FIG. 7 is a schematic diagram of an edge image obtained by synthesizing three edge sub-images in a specific embodiment
- FIG. 8 is a schematic diagram of an edge image after binarizing a synthesized edge image in a specific embodiment
- FIG. 9 is a schematic diagram of an edge image obtained by performing interference filtering on a binarized edge image in a specific embodiment
- Figure 10 is a schematic illustration of the position of a peak in a particular embodiment
- FIG. 11 is a schematic flow chart of a method for obtaining a color value of a color circle in an embodiment
- FIG. 12 is a schematic structural view of a color ring resistance detecting device in an embodiment
- FIG. 13 is a schematic structural diagram of a color circle candidate region determining module in an embodiment
- FIG. 14 is a schematic structural diagram of a color circle abscissa determining unit in an embodiment
- FIG. 15 is a schematic structural diagram of a color ring color value obtaining module in one embodiment.
- a method for detecting color ring resistance includes the following steps:
- This embodiment can be implemented by a corresponding program, and the program can be run in the automatic optical detection system.
- the color ring resistance is detected, it is no longer necessary to identify the auxiliary device by means of additional color ring resistance, and no additional arrangement of workers is required.
- the detection saves the hardware cost and the labor cost effectively, and improves the detection speed of the color ring resistance.
- the specific embodiments of the various steps are described in detail below.
- step S110 in order to realize the automatic detection of the color ring resistance, it is first necessary to extract an image of excellent ring resistance from the image of the circuit board, that is, to locate the main body area of the color ring resistance, and the main body area is the entire area of the color ring resistance.
- the circuit board to be tested is a circuit board that is actually produced and needs to be tested for defects.
- the image of the board to be inspected can be automatically scanned according to the camera carried by the automatic optical detection system itself.
- Extracting the image of the color ring resistance from the image of the circuit board to be inspected has various implementations.
- the image of the entire circuit board can be imaged according to the template image of the color ring resistance saved during the board plate making process. Matching to obtain an image of the color ring resistance, but this method is computationally intensive, especially when the number of color ring resistors is large. This method is inefficient.
- the color ring resistance is extracted from the image of the circuit board.
- the steps of the image may include:
- S1101 Obtain a preliminary position of the color ring resistance in an image of the circuit board according to a position of the color ring resistance set in the circuit board when the circuit board is plated;
- the position of the color ring resistor in the board is set.
- the actual board should be the same as the board designed during the plate making, so the color ring can be firstly made according to the board.
- the position of the resistor finds the position of the color ring resistor in the image of the board to be tested.
- the color ring resistance will be offset in the horizontal direction. For example, if the pin on the left side is inserted more, the color ring resistance will be on the left. Offset. Therefore, considering the entry and exit of the actual production circuit board and the plate-making circuit board, after finding the position of the color ring resistance in the image of the circuit board to be detected, the position is extended outward by a preset pixel, for example, 30-60 pixels are extended outward. Etc., the expanded region is used as a candidate region for the color ring resistance, where it is expanded outward to expand in various directions.
- S1103 Perform image matching on the candidate region of the color ring resistance according to the image of the color ring resistance in the plate making process, and obtain an image of the color ring resistance in the candidate region of the color ring resistance;
- the user saves the template image of the color ring resistance when performing the board plate making, and uses the template image of the color ring resistance to perform the image matching method to locate the main body area of the color ring resistance, and extract the image of the excellent ring resistance.
- the method of image matching can be implemented in accordance with the existing methods in the prior art.
- step S120 after obtaining the body region of the color ring resistor, the color ring region positioning of the color ring resistance is also required.
- the step of determining a candidate region of the color ring included in the color ring resistance according to the image of the color ring resistance may include:
- S1201 Determine an abscissa of a color ring included in the color ring resistance according to an image of the color ring resistance;
- step S1201 various implementations of obtaining the color circle abscissa by the image processing method, for example, in one embodiment, as shown in FIG. 4, determining the color ring resistance according to the image of the color ring resistance
- the steps of the abscissa of the color circle may include:
- the main area I of the color ring resistance is based on three colors of RGB.
- the channels are separated to obtain sub-images of I_R, I_G, and I_B.
- a sub-image diagram of each channel wherein the sub-image corresponding to the red channel and the green channel are sequentially from left to right.
- the corresponding sub-image and the sub-image corresponding to the blue channel are sequentially from left to right.
- the X-direction (horizontal direction) edge is performed on the sub-images of the three channels. Detection.
- edge detection of sub-images of each channel there are various implementations for edge detection of sub-images of each channel.
- the Sobel operator Sobel operator
- the edge sub-images G R , G G , G B of the R, G, and B channel images are obtained.
- the obtained edge sub-images of the respective channels are, from left to right, edge sub-images corresponding to the red channel, edge sub-images corresponding to the green channel, and edge sub-images corresponding to the blue channel.
- the step of synthesizing the first edge image according to the color circle edge detection result of all the sub-images may include: obtaining a maximum value of the same color ring in the color circle edge detection result of all the sub-images; according to the obtained respective colors The maximum value corresponding to the ring is combined to form a first edge image.
- G(x,y) max(G R (x,y), G G (x,y), G B (x,y)) (2)
- the synthesized edge image G is as shown in FIG.
- the edge image G is binarized by the binarization method, and the binarized edge image G is as shown in FIG.
- the method further includes the step of: binarizing the first edge image with two in the horizontal direction
- the value of a point whose edge distance is smaller than the set range is set to 0.
- the value at the two edges in the horizontal direction of the binarized edge image G is set to 0 to perform interference filtering to obtain a binarized image B.
- the two edges in the horizontal direction are the left and right edges of the binarized edge image G.
- the setting range can be set as needed.
- Mask can be used to remove interference values near the left and right edges.
- Binary value after zeroing The image B is as shown in FIG. 9. As can be seen from FIG. 9, there is no more interference at the left edge of the binarized image B that does not belong to the color ring region.
- the binarized edge image G is the second edge image, and the edge image G is subsequently summed in the Y direction.
- the binarized image B is the second edge image, and the binarized image B is subsequently summed in the Y direction.
- the pixel points in the second edge image are summed in the Y direction to obtain a vector B_X.
- summing the second edge images in the Y direction that is, summing the respective ordinates corresponding to each abscissa in the second edge image, for example, summing the respective ordinates corresponding to the abscissa 1 to obtain a horizontal
- the sum corresponding to the coordinate 1 is summed with the respective ordinates corresponding to the abscissa 2 to obtain the sum corresponding to the abscissa 2, and so on, and the sum of the ordinates corresponding to the respective abscissas in the second edge image is obtained.
- the sum of the ordinates corresponding to these abscissas constitutes a vector B_X.
- the method may further include the step of: smallizing the first vector The value of the preset threshold is set to 0.
- the preset threshold can be set as needed.
- the preset threshold is 255*H/2, where H is the height of the image of the color ring resistance, and each value in the vector B_X is judged and processed according to the following formula (3):
- the peak position is determined directly from the first vector obtained by summing in the Y direction. If the processing of the formula (3) is performed in the first vector, the peak position is determined based on the processed first vector.
- the first vector When determining the position of the peak, the first vector can be displayed in a histogram, and the abscissa of the color circle should be at the peak of the histogram, so the X coordinate of the peak is taken as the abscissa (X_color) of the color ring of the color ring resistance.
- FIG. 10 which is a schematic diagram of a specific embodiment of a histogram of a first vector, there are four peaks in FIG. 10, corresponding to four color rings in the color ring resistance, and the abscissa corresponding to each peak is the corresponding color. The abscissa of the ring.
- the present invention is not limited to the above manner of determining the horizontal axis of the color circle. Those skilled in the art may add other embodiments based on the above manner, for example, steps of increasing image sharpness processing, etc., The order of some steps is adjusted or an equivalent substitution is made for certain steps.
- step S1202 and step S1203 after obtaining the abscissa of the color circle, the preset length X_length is extended in the horizontal direction with the abscissa of the color circle as a center point, and the width of the color ring is determined.
- the horizontal direction is the left-right direction, and is expanded in the horizontal direction, that is, left and right, respectively.
- the method may further include: step according to the preset color circle width and the preset two adjacent color circles.
- the distance between the determinations determines the preset length.
- the preset length X_length may be determined according to a preset color ring width color_length and a distance between the two adjacent adjacent color rings. The type and position of the color ring resistor during board plate making have been determined, and the corresponding color ring width and the distance between adjacent color rings can be determined according to the determined color ring resistance model.
- the widths of the color rings of the color ring resistance are approximately equal, the distance between each adjacent two color rings is approximately equal, so the width of one color ring and the distance between any adjacent color rings are arbitrarily selected as a preset.
- step S130 after obtaining the candidate region of the color circle, the color information of the color circle can be obtained according to the candidate region. Since the color of the color of the colored ring resistor may also exist inside the candidate region of the color ring, and the color of each pixel on the color ring may also be slightly deviated due to noise, in order to ensure the accuracy of the obtained color value of the color ring, in one embodiment, as shown in FIG. 11, the color region of the candidate region of the color circle is obtained, and the step of obtaining the color value of the color ring may include:
- S1301 averaging values of points in the candidate region of the color circle in the Y direction to obtain a second vector; each of the second vectors is averaged in a Y direction corresponding to each abscissa value;
- the candidate region ROI_color of the obtained color circle is averaged in the Y direction (vertical direction), that is, the ordinates corresponding to each abscissa of the candidate region of the color circle are averaged, and each horizontal coordinate is obtained.
- the average of the colors, the average of the colors corresponding to the respective abscissas constitutes the vector vector_color.
- averaging the respective ordinates corresponding to the abscissa 1 of the candidate region to obtain an average value of the color corresponding to the abscissa 1, and the abscissa 2 of the candidate region The corresponding ordinates are averaged to obtain The average of the colors corresponding to the abscissa 2, and so on, obtains the average of the colors corresponding to the respective abscissas in the candidate region, that is, obtains the vector of the color circle.
- S1302 Compare each value in the second vector with a preset color value, and determine a preset color value when the difference is the smallest as a color value of a corresponding value in the second vector;
- Each value in the vector vector_color is compared with a preset color value, and the color with the smallest difference in color values is used as the color of the corresponding position of the vector.
- the color ring resistance generally has 12 colors
- the first value in the vector vector_color is compared with each color value in Table 1, respectively, and the color value and the first value in Table 1 are found.
- the color value with the smallest difference, then the color value found in Table 1 is the color value of the first value in the vector, and so on, find the color value of each value in the vector.
- Table 1 HSV (Hue, Saturation, Value) values of the color circle color in the color ring resistance
- colour HSV value colour HSV value Black (0,0,0) Green (120,100,50) Brown (0,75,65) Blue (240,100,100) Red (0,100,100) Purple (300,100,50) Orange (39,100,100) Gray (0,0,50) Yellow (60,100,100) White (0,0,100) Golden (51,100,100) Silver (0,0,75)
- S1303 Filter out, in the second vector, a color value that is the most frequently occurring and is not the color of the color of the color ring resistance, and uses the filtered color value as a color value of the color ring, and the main body color is a color removal.
- the color value of each value on the vector vector_color is excluded. After the color value of the color ring resistance body is excluded, the color with the most statistical value of the color value on the vector_color is taken as the color of the color circle.
- the color of the main body of the color ring resistor is the color printed on the surface of the color ring resistor. When the board is plated, the type, color and the like of the color ring resistor used can be determined.
- the wrong component detection of the color ring resistance refers to determining whether the resistance value of the color ring resistance is a predetermined size, and if it is not the resistance value of the target, it can be regarded as a wrong component.
- the reverse component detection of the color ring resistance refers to determining whether the color ring sequence of the color ring resistance is opposite to the specified order, and if it is the opposite, it can be considered as the reverse component. So, in one implementation In an example, the step of detecting the color ring resistance according to the color value of the color ring may include:
- the color ring resistance of the suspected defect After detecting the color ring resistance to be detected in the circuit board, the color ring resistance of the suspected defect can be displayed or marked according to the detection result, which is convenient for the user to view and repair.
- the present invention also provides a color ring resistance detecting device, and a specific embodiment of the device of the present invention will be described in detail below with reference to the accompanying drawings.
- a color ring resistance detecting device includes:
- the color ring resistance image extraction module 110 is configured to acquire an image of the circuit board to be detected, and extract an image of the color ring resistance from the image of the circuit board;
- a color circle candidate region determining module 120 configured to determine, according to the image of the color ring resistance, a candidate region of a color ring included in the color ring resistance;
- a color ring color value obtaining module 130 configured to perform color recognition on a candidate area of the color ring to obtain a color value of the color ring;
- the color ring resistance detecting module 140 is configured to detect the color ring resistance according to the color value of the color ring.
- the device of the embodiment can be operated in the automatic optical detection system, and when the color ring resistance is detected, it is no longer necessary to identify the auxiliary device by means of additional color ring resistance, and there is no need to arrange additional workers for detection, thereby effectively saving. Hardware and labor costs increase the detection speed of the color ring resistance.
- the functions of the respective modules will be described in detail below.
- the color ring resistance image extraction module 110 extracts images of the color ring resistance from the image of the circuit board to be detected in various ways. Considering the magnitude of the calculation amount and the deviation between the actually produced circuit board and the circuit board designed when the board is plated, in one embodiment, the color ring resistance image extraction module 110 may include:
- a preliminary position obtaining unit 1101 configured to obtain a preliminary position of the color ring resistance in an image of the circuit board according to a position of the color ring resistance set in the circuit board when the circuit board is plated;
- a color ring resistance candidate region obtaining unit 1102 configured to expand a preliminary position of the color ring resistor to each of the preset pixels to obtain a candidate region of the color ring resistance
- a color ring resistance image extracting unit 1103 configured to perform image matching on a candidate region of the color ring resistance according to an image of the color ring resistance in the board plate making process, to obtain the candidate region of the color ring resistance An image of the color ring resistance.
- the color circle candidate region determining module 120 may include:
- a color circle abscissa determining unit 1201, configured to determine an abscissa of a color ring included in the color ring resistance according to the image of the color ring resistance;
- a color circle width determining unit 1202 configured to expand a preset length in a horizontal direction with the abscissa of the color ring as a center point, to obtain a width of the color ring;
- the preset length X_length may be determined according to a preset color ring width color_length and a distance between the two adjacent adjacent color rings.
- the color circle candidate region determining unit 1203 is configured to obtain a candidate region of the color ring included in the color ring resistance according to the width of the color ring and the height of the image of the color ring resistance.
- the color circle abscissa determining unit 1201 may include:
- a sub-image obtaining sub-unit 1201a configured to convert an image of the color ring resistance into a sub-image corresponding to each color channel according to an RGB color mode
- the first edge image obtaining sub-unit 1201b is configured to perform color ring edge detection in the X direction for each sub image, and synthesize the first edge image according to the color ring edge detection result of all the sub images;
- a second edge image obtaining sub-unit 1201c configured to binarize the first edge image to obtain a second edge Edge image
- a first vector obtaining sub-unit 1201d for summing values of points in the second edge image in the Y direction to obtain a first vector; each of the first vectors is a value of each abscissa The value of the corresponding Y direction summation;
- the abscissa determining subunit 1201e is configured to determine a peak position in the second edge image from the first vector, and determine an X coordinate of the peak position as an abscissa of a color ring included in the color ring resistance.
- color circle abscissa determining unit 1201 is not limited to the above subunits, and those skilled in the art may add other subunits based on the color ring abscissa determining unit 1201, for example, adding an image sharpness processing subroutine. Units, etc., can also be equivalent to some subunits.
- the color ring color value obtaining module 130 can obtain the color information of the color ring according to the candidate region. Since the color of the color of the colored ring resistor may also exist inside the candidate region of the color ring, and the color of each pixel on the color ring may also be slightly deviated due to noise, in order to ensure the accuracy of the obtained color value of the color ring, in an embodiment, as shown in FIG. 15, the color ring color value obtaining module 130 may include:
- a second vector obtaining unit 1301, configured to average the values of the points in the candidate region of the color circle in the Y direction to obtain a second vector; each of the second vectors is a value of each abscissa The average value of the corresponding Y direction;
- the color value comparison unit 1302 is configured to compare each of the second vectors with a preset color value, and determine a preset color value when the difference is the smallest as a corresponding value in the second vector.
- a color circle color value determining unit 1303, configured to filter out a color value of a body color having the most occurrences in the second vector and not being the color ring resistance, and using the filtered color value as a color value of the color ring
- the body color is a color of a color ring resistance surface other than the color of the color ring.
- color ring resistance detecting device of the present invention is the same as those of the color ring resistance detecting method of the present invention, and are not described herein.
- the present invention also provides an automatic optical detection system comprising the color ring resistance detecting device of any of the above embodiments.
- the method and device for detecting the color ring resistance and the automatic optical detecting system automatically scan the circuit board to acquire an image, and then extract an image of the color ring resistance, and further accurately extract the color ring included in the excellent ring resistance through image processing technology.
- the color value so that the color ring resistance is judged according to the color value of each color ring, whether there is a wrong or reverse defect, the automatic detection of the color ring resistance is realized, and the auxiliary device is no longer needed by the additional color ring resistance, and no additional Arranging workers for testing effectively saves hardware and labor costs.
- the speed of the color ring resistance detection is no longer limited by the skill of the worker, and the detection speed of the color ring resistance is greatly improved.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
一种色环电阻的检测方法、装置和自动光学检测系统。所述方法包括步骤：获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像(S110)；根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域(S120)；对所述色环的候选区域进行颜色识别，获得所述色环的颜色值(S130)；根据所述色环的颜色值对所述色环电阻进行检测(S140)。该方法有效节省了硬件成本和人工成本，大大提高了色环电阻的检测速度。
Description
本发明涉及电子技术领域，特别是涉及一种色环电阻的检测方法、色环电阻的检测装置和自动光学检测系统。
自动光学检测(AOI，Automated Optical Inspection)是工业制作过程的必要环节，已经广泛应用于TFT-LCD(thin film transistor-liquid crystal display，薄膜晶体管液晶显示器)、晶体管与PCB(Printed Circuit Board，印制电路板)等工业制作上，在民生用途则延伸至保全系统。它使用机器视觉作为检测标准技术，利用光学方式取得成品的表面状态，以影像处理来检测异物或表面瑕疵。
电子元件的错、漏和反检测是电路板缺陷检测领域中的一种常见应用。作为电路板中常用的电子元件，色环电阻的错、漏、反检测极为重要。传统技术中AOI系统对色环电阻的缺陷检测一般集中在漏件检测上，而对色环电阻的错件和反件检测则需要人工借助额外的色环电阻识别辅助设备配合进行，如色环电阻直读卡、色环电阻识别仪等，这种方案需要耗费额外的硬件与人力，成本较高。
发明内容
基于此，有必要针对上述问题，提供一种低成本的色环电阻的检测方法、装置和自动光学检测系统。
为了达到上述目的，本发明采取的技术方案如下：
一种色环电阻的检测方法，包括步骤：
获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；
根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；
对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；
根据所述色环的颜色值对所述色环电阻进行检测。
上述色环电阻的检测方法，自动扫描电路板获取图像，然后提取色环电阻的图像，并
通过图像处理技术，进一步提取色环电阻包含的色环的颜色值，从而根据各个色环的颜色值判断色环电阻是否存在错、反等缺陷，实现了色环电阻的自动检测，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了硬件成本和人工成本。另外，由于不再需要人工对色环电阻进行检测，所以色环电阻检测的速度不再受限于工人的熟练程度，大大提高了色环电阻的检测速度。
在一个实施例中，根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域的步骤包括：根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包含的色环的候选区域。根据提取的色环电阻的图像自动定位色环电阻包含的色环的区域，准确性较高。
在一个实施例中，以所述色环的横坐标为中心点沿水平方向各扩展预设长度之前，还包括步骤：根据预设的色环宽度以及预设的相邻两个色环之间的距离确定所述预设长度。根据预设的色环宽度以及预设的相邻两个色环之间的距离确定色环的横坐标需要扩展的长度，进而可以准确获得色环的宽度。
在一个实施例中，根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标的步骤包括：将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；对所述第一边缘图像进行二值化，获得第二边缘图像；对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标。考虑到某些色环在灰度图像与色环电阻的主体颜色的灰色度相似，采用分通道边缘检测的方法，有效提高了色环横坐标确定的准确性。
在一个实施例中，根据全部子图像的色环边缘检测结果合成第一边缘图像的步骤包括：获得同一色环在全部子图像的色环边缘检测结果中的最大值；根据获得的各个色环对应的最大值合成第一边缘图像。将分通道边缘检测后的图像按照最大值进行融合，后续根据融合的图像确定色环的横坐标，有效提高了色环横坐标确定的准确性。
在一个实施例中，对所述第一边缘图像进行二值化之后，获得第二边缘图像之前，还
包括步骤：将二值化后的所述第一边缘图像中与水平方向的两个边缘距离小于设定范围的点的值设置为0。考虑到色环不会出现在色环电阻的左右边缘处，将二值化后的图像左右边缘附近的值设置为0，从而有效过滤干扰信息，有效提高了色环横坐标确定的准确性。
在一个实施例中，得到第一向量之后，由第一向量确定所述第二边缘图像中的波峰位置之前，还包括步骤：将所述第一向量中小于预设阈值的值设置为0。对第一向量中小于预设阈值的值设置为0，从而能够更为准确的定位出色环所对应的波峰位置。
在一个实施例中，对所述色环的候选区域进行颜色识别，获得所述色环的颜色值的步骤包括：对所述色环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环颜色之外的色环电阻表面的颜色。考虑到色环候选区域还可能存在着色环电阻的主体颜色，而且色环上的每个像素点的颜色也可能因为噪声有点偏差，所以对获得的色环的候选区域颜色值在纵坐标方向上求平均，从而根据颜色的平均值确定色环的颜色，提高了色环颜色值确定的准确性。
在一个实施例中，从所述电路板的图像中提取色环电阻的图像的步骤包括：根据所述电路板制版时设定的色环电阻在所述电路板中的位置，获得所述电路板的图像中所述色环电阻的初步位置；将所述色环电阻的初步位置向外各扩展预设像素，获得所述色环电阻的候选区域；根据所述电路板制版时所述色环电阻的图像，对所述色环电阻的候选区域进行图像匹配，获得所述色环电阻的候选区域中所述色环电阻的图像。考虑到实际生产出的电路板与制版时电路板的偏差，对生产电路板的图像中色环电阻的位置进行扩展，从而得到色环电阻的候选区域，根据该候选区域提取色环电阻的图像，大大提高了色环电阻图像提取的效率和准确度。
一种色环电阻的检测装置，包括：
色环电阻图像提取模块，用于获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；
色环候选区域确定模块，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；
色环颜色值获得模块，用于对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；
色环电阻检测模块，用于根据所述色环的颜色值对所述色环电阻进行检测。
上述色环电阻的检测装置，自动扫描电路板获取图像，然后提取色环电阻的图像，并通过图像处理技术，进一步提取色环电阻包含的色环的颜色值，从而根据各个色环的颜色值判断色环电阻是否存在错、反等缺陷，实现了色环电阻的自动检测，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了硬件成本和人工成本。另外，由于不再需要人工对色环电阻进行检测，所以色环电阻检测的速度不再受限于工人的熟练程度，大大提高了色环电阻的检测速度。
在一个实施例中，所述色环候选区域确定模块包括：色环横坐标确定单元，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；色环宽度确定单元，用于以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；色环候选区域确定单元，用于根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包含的色环的候选区域。所述色环候选区域确定模块根据提取的色环电阻的图像自动定位色环电阻包含的色环的区域，准确性较高。
在一个实施例中，所述色环横坐标确定单元包括：子图像获得子单元，用于将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；第一边缘图像获得子单元，用于对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；第二边缘图像获得子单元，用于对所述第一边缘图像进行二值化，获得第二边缘图像；第一向量获得子单元，用于对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；横坐标确定子单元，用于由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标。考虑到某些色环在灰度图像与色环电阻的主体颜色的灰色度相似，采用分通道边缘检测的方法，有效提高了色环横坐标确定的准确性。
在一个实施例中，所述色环颜色值获得模块包括：第二向量获得单元，用于对所述色
环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；颜色值比较单元，用于将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；色环颜色值确定单元，用于筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环颜色之外的色环电阻表面的颜色。考虑到色环候选区域还可能存在着色环电阻的主体颜色，而且色环上的每个像素点的颜色也可能因为噪声有点偏差，所以所述色环颜色值获得模块对获得的色环的候选区域颜色值在纵坐标方向上求平均，从而根据颜色的平均值确定色环的颜色，提高了色环颜色值确定的准确性。
一种自动光学检测系统，包括上述任意一项所述的色环电阻的检测装置。该自动光学检测系统可以自动扫描电路板获取图像，然后提取色环电阻的图像，并通过图像处理技术，进一步提取色环电阻包含的色环的颜色值，从而根据各个色环的颜色值判断色环电阻是否存在错、反等缺陷，实现了色环电阻的自动检测，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了硬件成本和人工成本。另外，由于不再需要人工对色环电阻进行检测，所以色环电阻检测的速度不再受限于工人的熟练程度，大大提高了色环电阻的检测速度。
图1为一个实施例中色环电阻的检测方法的流程示意图；
图2为一个实施例中提取色环电阻图像方法的流程示意图；
图3为一个实施例中确定色环候选区域方法的流程示意图；
图4为一个实施例中确定色环的横坐标方法的流程示意图；
图5为一个具体实施例中根据RGB划分的三个子图像的示意图；
图6为一个具体实施例中对三个子图像进行水平方向的边缘检测后得到的三个边缘子图像的示意图；
图7为一个具体实施例中对三个边缘子图像进行合成后得到的边缘图像的示意图；
图8为一个具体实施例中对合成的边缘图像进行二值化后的边缘图像的示意图；
图9为一个具体实施例中对二值化后的边缘图像进行干扰过滤后得到的边缘图像的示意图；
图10为一个具体实施例中波峰位置的示意图；
图11为一个实施例中获得色环颜色值的方法的流程示意图；
图12为一个实施例中色环电阻的检测装置的结构示意图；
图13为一个实施例中色环候选区域确定模块的结构示意图；
图14为一个实施例中色环横坐标确定单元的结构示意图；
图15为一个实施例中色环颜色值获得模块的结构示意图。
为更进一步阐述本发明所采取的技术手段及取得的效果，下面结合附图及较佳实施例，对本发明的技术方案，进行清楚和完整的描述。
如图1所示，一种色环电阻的检测方法，包括步骤：
S110、获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；
S120、根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；
S130、对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；
S140、根据所述色环的颜色值对所述色环电阻进行检测。
本实施例可以通过相应的程序实现，程序可以运行在自动光学检测系统中，则在对色环电阻进行检测时，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了硬件成本和人工成本，提高了色环电阻的检测速度。为了更好地理解本发明，下面对各个步骤的具体实施方式做详细描述。
在步骤S110中，为了实现色环电阻的自动检测，首先需要从电路板的图像中提取出色环电阻的图像，即对色环电阻的主体区域进行定位，主体区域即为色环电阻的整个区域。待检测的电路板为实际生产出来的需要进行缺陷检测的电路板。待检测电路板的图像可以根据自动光学检测系统自身携带的摄像头自动扫描获得。
从待检测电路板的图像中提取色环电阻的图像有多种实现方式，例如，在一个实施例中，可以根据所述电路板制版时保存的色环电阻的模板图像对整个电路板的图像进行匹配，从而获得色环电阻的图像，但是该种方法计算量较大，尤其在色环电阻数量很大时，
该种方法效率很低。
因此，考虑计算量的大小以及实际生产的电路板与电路板制版时设计的电路板的偏差，在另一个实施例中，如图2所示，从所述电路板的图像中提取色环电阻的图像的步骤可以包括：
S1101、根据所述电路板制版时设定的色环电阻在所述电路板中的位置，获得所述电路板的图像中所述色环电阻的初步位置；
用户对电路板制版时，会设定好色环电阻在该电路板中的位置，理想状态下，实际生产的电路板应该与制版时设计的电路板相同，所以可以先根据电路板制版时色环电阻的位置找到色环电阻在待检测电路板图像中的位置。
S1102、将所述色环电阻的初步位置向外各扩展预设像素，获得所述色环电阻的候选区域；
如果色环电阻两边引脚插入插孔的长度不一样，那么就会使色环电阻在水平方向上出现偏移，比如左边的插孔引脚插入的比较多，那么色环电阻就会向左边偏移。因此，考虑到实际生产电路板与制版电路板的出入，找到色环电阻在待检测电路板图像中的位置后，将该位置向外扩展预设个像素，例如向外扩展30-60个像素等，将扩展后的区域作为色环电阻的候选区域，其中向外扩展为向各个方向扩展。
S1103、根据所述电路板制版时所述色环电阻的图像，对所述色环电阻的候选区域进行图像匹配，获得所述色环电阻的候选区域中所述色环电阻的图像；
用户在进行电路板制版时会保存有色环电阻的模板图像，利用色环电阻的模板图像，采用图像匹配的方法进行色环电阻的主体区域定位，提取出色环电阻的图像。图像匹配的方法可以根据现有技术中已有的方式实现。
在步骤S120中，获得色环电阻的主体区域后，还需要进行色环电阻的色环区域定位。在一个实施例中，如图3所示，根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域的步骤可以包括：
S1201、根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；
S1202、以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；
S1203、根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包
含的色环的候选区域。
为了更好的理解色环的候选区域确定的方式，下面对上述三个步骤分别进行详细介绍。
在步骤S1201中，通过图像处理方法获得色环横坐标有多种实现方式，例如，在一个实施例中，如图4所示，根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标的步骤可以包括：
S1201a、将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；
因为某些色环在灰度图像与色环电阻的主体颜色的灰度值相似，无法很好的区分，因此需要充分的考虑颜色信息，将色环电阻的主体区域I，根据RGB三种颜色通道分开，得到I_R、I_G、I_B的子图像。例如，如图5所示，为将色环电阻的主体区域按照RGB模式分为三个颜色通道后，各个通道的子图像示意图，其中从左到右依次为红通道对应的子图像、绿通道对应的子图像以及蓝通道对应的子图像。
S1201b、对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；
因为色环电阻的色环是垂直印刷在色环电阻的主体上的，因此，在获得色环电阻三个通道的子图像后，对三个通道的子图像进行X方向(水平方向)的边缘检测。
对各通道的子图像进行边缘检测有多种实现方式，例如，利用Sobel算子(Sobel operator，索贝尔算子)对各通道的子图像进行X方向的边缘检测，也即是利用式(1)得到R、G、B通道图像的边缘子图像GR、GG、GB。如图6所示，为得到的各个通道的边缘子图像，从左到右依次为红通道对应的边缘子图像、绿通道对应的边缘子图像以及蓝通道对应的边缘子图像。
分通道进行边缘检测后，还需要对边缘检测后的图像进行融合。在一个实施例中，根据全部子图像的色环边缘检测结果合成第一边缘图像的步骤可以包括：获得同一色环在全部子图像的色环边缘检测结果中的最大值；根据获得的各个色环对应的最大值合成第一边缘图像。
获得RGB三个通道的X方向的边缘子图像后，我们取对应位置的最大值，作为合成的边缘图像G，如式(2)所示。
G(x,y)＝max(GR(x,y),GG(x,y),GB(x,y)) (2)
合成后的边缘图像G如图7所示。
S1201c、对所述第一边缘图像进行二值化，获得第二边缘图像；
获得了合成后的边缘图像G后，再利用二值化方法对边缘图像G进行二值化，二值化后的边缘图像G如图8所示。
从图8可以看出，在二值化后的边缘图像G左边缘附近(矩形虚线框内)有些不属于色环区域的干扰，而实际情况中色环并不会出现在左右边缘处。因此，在一个实施例中，对所述第一边缘图像进行二值化之后，获得第二边缘图像之前，还包括步骤：将二值化后的所述第一边缘图像中与水平方向的两个边缘距离小于设定范围的点的值设置为0。将二值化后的边缘图像G中水平方向的两个边缘处的值设置为0，以进行干扰过滤，得到二值化图像B。水平方向的两个边缘即为二值化后的边缘图像G左边边缘和右边边缘。设定范围可以根据需要进行设定。
在一个具体实施例中，可以利用Mask将靠近左右边缘处的干扰值去掉。例如，可以设定mask_length＝color_length长度的区域全部值为0，其中，mask_length为从二值化后的边缘图像G左边边缘或右边边缘到二值化后的边缘图像G中心的长度。置零后的二值
化图像B如图9所示，从图9可以看出，二值化图像B左边缘处不再存在不属于色环区域的干扰。
需要说明的是，如果不进行边缘干扰处理，二值化后的边缘图像G为第二边缘图像，后续对边缘图像G进行Y方向的求和。如果进行边缘干扰处理，则二值化图像B为第二边缘图像，后续对该二值化图像B进行Y方向的求和。
S1201d、对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；
在获得了第二边缘图像后，对第二边缘图像中的像素点在Y方向进行求和，得到向量B_X。对第二边缘图像在Y方向进行求和，即对第二边缘图像中每一个横坐标所对应的各个纵坐标进行求和，例如对横坐标1所对应的各个纵坐标进行求和，得到横坐标1所对应的和，对横坐标2所对应的各个纵坐标进行求和，得到横坐标2所对应的和，依次类推，获得第二边缘图像中各个横坐标所对应的纵坐标的和，这些横坐标对应的纵坐标的和组成一个向量B_X。
为了更为准确的确定波峰位置，在一个实施例中，得到第一向量之后，由第一向量确定所述第二边缘图像中的波峰位置之前，还可以包括步骤：将所述第一向量中小于预设阈值的值设置为0。
预设阈值可以根据需要进行设定。例如预设阈值为255*H/2，H为色环电阻的图像的高度，对向量B_X中的每个值按照下式(3)进行判断和处理：
S1201e、由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标；
如果不进行公式(3)的处理步骤，则直接根据Y方向求和得到的第一向量确定波峰位置。如果在第一向量进行了公式(3)的处理，则根据处理后的第一向量确定波峰位置。
在确定波峰位置时，可以将第一向量以直方图显示，色环的横坐标应该在直方图的波峰处，因此将波峰的X坐标作为色环电阻的色环的横坐标(X_color)。如图10所示，为第一向量的直方图的具体实施例的示意图，图10中有四个波峰，对应色环电阻中的四个色环，每一个波峰对应的横坐标即为对应色环的横坐标。
需要说明的是，本发明并不限制于上述确定色环横坐标的方式，本领域技术人员可以在上述方式的基础上增加其他实施方式，例如，增加图像清晰度处理的步骤等，也可以对某些步骤的顺序进行调整或者对某些步骤进行等同替代。
在步骤S1202和步骤S1203中，获得色环的横坐标后，以所述色环的横坐标为中心点沿水平方向各扩展预设长度X_length，确定色环的宽度。水平方向即左右方向，沿水平方向扩展即左右分别进行扩展。
在一个实施例中，以所述色环的横坐标为中心点沿水平方向各扩展预设长度之前，还可以包括步骤：根据预设的色环宽度以及预设的相邻两个色环之间的距离确定所述预设长度。所述预设长度X_length可以根据预设的色环宽度color_length以及预设的相邻两个色环之间的距离distance确定。电路板制版时色环电阻的型号和位置等已经确定，根据确定的色环电阻的型号即可以确定出对应的色环宽度以及相邻色环之间的距离。由于色环电阻的各个色环的宽度近似相等，每相邻两个色环之间的距离近似相等，所以任意选择一个色环的宽度和任意一个相邻色环之间的距离作为预设的色环宽度以及预设的相邻两个色环之间的距离。
在一个具体实施例中，可以根据X_length＝color_length/2+distance/3确定色环宽度。获得色环的宽度后，根据色环电阻的图像的高度，即确定出色环的候选区域。
在步骤S130中，获得色环的候选区域后，就可以根据该候选区域得到色环的颜色信息。由于色环的候选区域内部还可能存在着色环电阻的主体颜色，而且色环上的每个像素点的颜色也可能因为噪声有点偏差，因此，为了保证获得的色环颜色值的准确性，在一个实施例中，如图11所示，对所述色环的候选区域进行颜色识别，获得所述色环的颜色值的步骤可以包括：
S1301、对所述色环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；
对获得的色环的候选区域ROI_color在Y方向(竖直方向)上求平均，即对色环的候选区域中每一个横坐标所对应的各个纵坐标进行求平均，获得每一个横坐标所对应的颜色的平均值，各个横坐标所对应的颜色的平均值构成向量vector_color。例如，获得一个色环的候选区域后，对该候选区域的横坐标1所对应的各个纵坐标进行求平均，得到横坐标1所对应的颜色的平均值，对该候选区域的横坐标2所对应的各个纵坐标进行求平均，得到
横坐标2所对应的颜色的平均值，以此类推，获得该候选区域中各个横坐标所对应的颜色的平均值，即获得该色环的向量。
S1302、将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；
分别将向量vector_color中的每个值与预设的各个颜色值进行比较，将颜色值差别最小的颜色作为该向量对应位置的颜色。例如，如表1所示，色环电阻一般有12种颜色，将向量vector_color中的第一个值与表1中的各个颜色值分别进行比较，找到表1中的颜色值与第一个值相差最小的颜色值，那么找到的表1中的颜色值即为向量中第一个值的颜色值，依次类推，找到向量中每一个值的颜色值。
表1 色环电阻中色环颜色的HSV(Hue,Saturation,Value)值
| 颜色 | HSV值 | 颜色 | HSV值 |
| Black | (0，0，0) | Green | (120，100，50) |
| Brown | (0，75，65) | Blue | (240，100，100) |
| Red | (0，100，100) | Purple | (300，100，50) |
| Orange | (39，100，100) | Gray | (0，0，50) |
| Yellow | (60，100，100) | White | (0，0，100) |
| Golden | (51，100，100) | Silver | (0，0，75) |
S1303、筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环颜色之外的色环电阻表面的颜色；
统计向量vector_color上每个值的颜色值，排除掉色环电阻主体颜色值后，取vector_color上颜色值统计次数最多的颜色作为色环的颜色。色环电阻的主体颜色即为色环电阻表面所印刷的颜色，在电路板制版时即可以确定出所采用的色环电阻的型号、颜色等等。
在步骤S140中，色环电阻的错件检测是指判断色环电阻的阻值是否是规定的大小，如果不是目标的阻值大小，那么即可认为是错件。色环电阻的反件检测是指判断色环电阻的色环顺序是否与规定的顺序相反，如果相反，那么即可认为是反件。所以，在一个实施
例中，根据所述色环的颜色值对所述色环电阻进行检测的步骤可以包括：
若所述色环电阻中各个色环的颜色值的顺序与预设的各个色环的颜色值的顺序相同，确定所述色环电阻为正常件；
若所述色环电阻中各个色环的颜色值的顺序与预设的各个色环的颜色值的顺序相反，确定所述色环电阻为反件；
若所述色环电阻中各个色环的颜色值的顺序与预设的各个色环的颜色值的顺序不相同且不相反，确定所述色环电阻为错件。
例如，待检测色环电阻包含四个色环，根据上述步骤确定出该色环电阻四个色环从左到右的颜色是color＝[color1，color2，color3，color4]。规定的颜色值为：set_color＝[set_color1，set_color2，set_color3，set_color4]。如果color1＝set_color1，color2＝set_color2，color3＝set_color3，color4＝set_color4，则待检测色环电阻为正常件。如果color1＝set_color4，color2＝set_color3，color3＝set_color2，color4＝set_color1，则待检测色环电阻为反件。如果待检测色环电阻不满足上述正常件和反件的判断条件，则确定为错件。
对电路板中待检测色环电阻检测后，可以根据检测的结果将疑似缺陷的色环电阻显示或标记出来，方便用户查看与检修。
基于同一发明构思，本发明还提供一种色环电阻的检测装置，下面结合附图对本发明装置的具体实施方式做详细描述。
如图12所示，一种色环电阻的检测装置，包括：
色环电阻图像提取模块110，用于获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；
色环候选区域确定模块120，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；
色环颜色值获得模块130，用于对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；
色环电阻检测模块140，用于根据所述色环的颜色值对所述色环电阻进行检测。
本实施例装置可以运行在自动光学检测系统中，则在对色环电阻进行检测时，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了
硬件成本和人工成本，提高了色环电阻的检测速度。为了更好地理解本发明，下面对各个模块的功能做详细描述。
色环电阻图像提取模块110从待检测电路板的图像中提取色环电阻的图像有多种实现方式。考虑计算量的大小以及实际生产的电路板与电路板制版时设计的电路板的偏差，在一个实施例中，所述色环电阻图像提取模块110可以包括：
初步位置获得单元1101，用于根据所述电路板制版时设定的色环电阻在所述电路板中的位置，获得所述电路板的图像中所述色环电阻的初步位置；
色环电阻候选区域获得单元1102，用于将所述色环电阻的初步位置向外各扩展预设像素，获得所述色环电阻的候选区域；
色环电阻图像提取单元1103，用于根据所述电路板制版时所述色环电阻的图像，对所述色环电阻的候选区域进行图像匹配，获得所述色环电阻的候选区域中所述色环电阻的图像。
获得色环电阻的主体区域后，还需要进行色环电阻的色环区域定位。在一个实施例中，如图13所示，所述色环候选区域确定模块120可以包括：
色环横坐标确定单元1201，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；
色环宽度确定单元1202，用于以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；
所述预设长度X_length可以根据预设的色环宽度color_length以及预设的相邻两个色环之间的距离distance确定。
色环候选区域确定单元1203，用于根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包含的色环的候选区域。
在一个实施例中，如图14所示，所述色环横坐标确定单元1201可以包括：
子图像获得子单元1201a，用于将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；
第一边缘图像获得子单元1201b，用于对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；
第二边缘图像获得子单元1201c，用于对所述第一边缘图像进行二值化，获得第二边
缘图像；
第一向量获得子单元1201d，用于对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；
横坐标确定子单元1201e，用于由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标。
需要说明的是，色环横坐标确定单元1201并不限制于上述子单元，本领域技术人员可以在上述色环横坐标确定单元1201的基础上增加其他子单元，例如，增加图像清晰度处理子单元等，也可以对某些子单元进行等同替代。
色环候选区域确定模块120获得色环的候选区域后，色环颜色值获得模块130就可以根据该候选区域得到色环的颜色信息。由于色环的候选区域内部还可能存在着色环电阻的主体颜色，而且色环上的每个像素点的颜色也可能因为噪声有点偏差，因此，为了保证获得的色环颜色值的准确性，在一个实施例中，如图15所示，所述色环颜色值获得模块130可以包括：
第二向量获得单元1301，用于对所述色环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；
颜色值比较单元1302，用于将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；
色环颜色值确定单元1303，用于筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环颜色之外的色环电阻表面的颜色。
本发明色环电阻的检测装置的其它技术特征与本发明色环电阻的检测方法的技术特征相同，在此不予赘述。
本发明还提供一种自动光学检测系统，所述自动光学检测系统包括上述任意一个实施例所述的色环电阻的检测装置。
上述色环电阻的检测方法、装置和自动光学检测系统，自动扫描电路板获取图像，然后提取色环电阻的图像，并通过图像处理技术，进一步准确提取出色环电阻包含的色环的
颜色值，从而根据各个色环的颜色值判断色环电阻是否存在错、反等缺陷，实现了色环电阻的自动检测，不再需要借助于额外的色环电阻识别辅助设备，也不需要额外安排工人进行检测，有效的节省了硬件成本和人工成本。另外，由于不再需要人工对色环电阻进行检测，所以色环电阻检测的速度不再受限于工人的熟练程度，大大提高了色环电阻的检测速度。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。
Claims (12)
- 一种色环电阻的检测方法，其特征在于，包括步骤：获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；根据所述色环的颜色值对所述色环电阻进行检测。
- 根据权利要求1所述的色环电阻的检测方法，其特征在于，根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域的步骤包括：根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包含的色环的候选区域。
- 根据权利要求2所述的色环电阻的检测方法，其特征在于，以所述色环的横坐标为中心点沿水平方向各扩展预设长度之前，还包括步骤：根据预设的色环宽度以及预设的相邻两个色环之间的距离确定所述预设长度。
- 根据权利要求2所述的色环电阻的检测方法，其特征在于，根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标的步骤包括：将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；对所述第一边缘图像进行二值化，获得第二边缘图像；对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标。
- 根据权利要求4所述的色环电阻的检测方法，其特征在于，根据全部子图像的色环边缘检测结果合成第一边缘图像的步骤包括：获得同一色环在全部子图像的色环边缘检测结果中的最大值；根据获得的各个色环对应的最大值合成第一 边缘图像；对所述第一边缘图像进行二值化之后，获得第二边缘图像之前，还包括步骤：将二值化后的所述第一边缘图像中与水平方向的两个边缘距离小于设定范围的点的值设置为0；得到第一向量之后，由第一向量确定所述第二边缘图像中的波峰位置之前，还包括步骤：将所述第一向量中小于预设阈值的值设置为0。
- 根据权利要求1所述的色环电阻的检测方法，其特征在于，对所述色环的候选区域进行颜色识别，获得所述色环的颜色值的步骤包括：对所述色环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环颜色之外的色环电阻表面的颜色。
- 根据权利要求1至6任意一项所述的色环电阻的检测方法，其特征在于，从所述电路板的图像中提取色环电阻的图像的步骤包括：根据所述电路板制版时设定的色环电阻在所述电路板中的位置，获得所述电路板的图像中所述色环电阻的初步位置；将所述色环电阻的初步位置向外扩展预设像素，获得所述色环电阻的候选区域；根据所述电路板制版时所述色环电阻的图像，对所述色环电阻的候选区域进行图像匹配，获得所述色环电阻的候选区域中所述色环电阻的图像。
- 一种色环电阻的检测装置，其特征在于，包括：色环电阻图像提取模块，用于获取待检测的电路板的图像，从所述电路板的图像中提取色环电阻的图像；色环候选区域确定模块，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的候选区域；色环颜色值获得模块，用于对所述色环的候选区域进行颜色识别，获得所述色环的颜色值；色环电阻检测模块，用于根据所述色环的颜色值对所述色环电阻进行检测。
- 根据权利要求8所述的色环电阻的检测装置，其特征在于，所述色环候选区域确定模块包括：色环横坐标确定单元，用于根据所述色环电阻的图像确定所述色环电阻包含的色环的横坐标；色环宽度确定单元，用于以所述色环的横坐标为中心点沿水平方向各扩展预设长度，获得所述色环的宽度；色环候选区域确定单元，用于根据所述色环的宽度以及所述色环电阻的图像的高度，获得所述色环电阻包含的色环的候选区域。
- 根据权利要求9所述的色环电阻的检测装置，其特征在于，所述色环横坐标确定单元包括：子图像获得子单元，用于将所述色环电阻的图像按照RGB颜色模式转换为各个颜色通道对应的子图像；第一边缘图像获得子单元，用于对各个子图像分别进行X方向的色环边缘检测，根据全部子图像的色环边缘检测结果合成第一边缘图像；第二边缘图像获得子单元，用于对所述第一边缘图像进行二值化，获得第二边缘图像；第一向量获得子单元，用于对所述第二边缘图像中的点的值在Y方向进行求和，得到第一向量；所述第一向量中的每一个值为每一个横坐标所对应的Y方向求和的值；横坐标确定子单元，用于由所述第一向量确定所述第二边缘图像中的波峰位置，将波峰位置的X坐标确定为所述色环电阻包含的色环的横坐标。
- 根据权利要求8至10任意一项所述的色环电阻的检测装置，其特征在于，所述色环颜色值获得模块包括：第二向量获得单元，用于对所述色环的候选区域中的点的值在Y方向上求平均，获得第二向量；所述第二向量中的每一个值为每一个横坐标所对应的Y方向求平均的值；颜色值比较单元，用于将所述第二向量中的每一个值分别与预设的颜色值进行比较，将差值最小时的预设的颜色值确定为所述第二向量中对应值的颜色值；色环颜色值确定单元，用于筛选出所述第二向量中出现次数最多且不为所述色环电阻的主体颜色的颜色值，将筛选出的颜色值作为所述色环的颜色值，所述主体颜色为除色环 颜色之外的色环电阻表面的颜色。
- 一种自动光学检测系统，其特征在于，包括权利要求8至11任意一项所述的色环电阻的检测装置。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610900096.6 | 2016-10-14 | ||
| CN201610900096.6A CN106529572B (zh) | 2016-10-14 | 2016-10-14 | 色环电阻的检测方法、装置和自动光学检测系统 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018068414A1 true WO2018068414A1 (zh) | 2018-04-19 |
Family
ID=58331860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/113145 WO2018068414A1 (zh) | 2016-10-14 | 2016-12-29 | 色环电阻的检测方法、装置和自动光学检测系统 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106529572B (zh) |
| WO (1) | WO2018068414A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351100A (zh) * | 2023-12-04 | 2024-01-05 | 成都数之联科技股份有限公司 | 一种色环电阻颜色提取方法、装置、设备和介质 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118433B (zh) * | 2017-06-22 | 2022-03-29 | 南京理工大学 | 一种色环电阻色码识别方法 |
| CN112419245A (zh) * | 2020-11-11 | 2021-02-26 | 广州视琨电子科技有限公司 | 色环电阻检测方法、装置、设备及存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001068900A (ja) * | 1999-08-30 | 2001-03-16 | Sony Corp | 実装部品検査方法並びに実装部品検査装置及びその照明器具 |
| CN102081802A (zh) * | 2011-01-26 | 2011-06-01 | 北京中星微电子有限公司 | 一种基于块匹配的色卡检测方法和装置 |
| CN105067629A (zh) * | 2015-07-24 | 2015-11-18 | 北京理工大学 | 一种色环电阻外观缺陷和阻值集成检测方法 |
| CN105957065A (zh) * | 2016-04-21 | 2016-09-21 | 广州视源电子科技股份有限公司 | 元件漏检方法和系统 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2385379Y (zh) * | 1999-07-16 | 2000-06-28 | 河南英威东风机械制造有限公司 | 色环电阻快速识别器 |
| CN103400366B (zh) * | 2013-07-03 | 2016-04-13 | 西安电子科技大学 | 基于条纹结构光的动态场景深度获取方法 |
| CN103530606B (zh) * | 2013-09-30 | 2016-06-29 | 中国农业大学 | 一种杂草环境下的农机导航路径提取方法 |
| CN105205802B (zh) * | 2015-02-13 | 2017-04-12 | 比亚迪股份有限公司 | 纹线距离的计算方法和装置 |
-
2016
- 2016-10-14 CN CN201610900096.6A patent/CN106529572B/zh active Active
- 2016-12-29 WO PCT/CN2016/113145 patent/WO2018068414A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001068900A (ja) * | 1999-08-30 | 2001-03-16 | Sony Corp | 実装部品検査方法並びに実装部品検査装置及びその照明器具 |
| CN102081802A (zh) * | 2011-01-26 | 2011-06-01 | 北京中星微电子有限公司 | 一种基于块匹配的色卡检测方法和装置 |
| CN105067629A (zh) * | 2015-07-24 | 2015-11-18 | 北京理工大学 | 一种色环电阻外观缺陷和阻值集成检测方法 |
| CN105957065A (zh) * | 2016-04-21 | 2016-09-21 | 广州视源电子科技股份有限公司 | 元件漏检方法和系统 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117351100A (zh) * | 2023-12-04 | 2024-01-05 | 成都数之联科技股份有限公司 | 一种色环电阻颜色提取方法、装置、设备和介质 |
| CN117351100B (zh) * | 2023-12-04 | 2024-03-22 | 成都数之联科技股份有限公司 | 一种色环电阻颜色提取方法、装置、设备和介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106529572A (zh) | 2017-03-22 |
| CN106529572B (zh) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108460757B (zh) | 一种手机TFT-LCD屏Mura缺陷在线自动检测方法 | |
| CN107945184B (zh) | 一种基于彩色图像分割和梯度投影定位的贴装元件检测方法 | |
| WO2017181724A1 (zh) | 电子元件漏件检测方法和系统 | |
| WO2018010391A1 (zh) | 一种板卡的检测方法及装置 | |
| US20070189606A1 (en) | Automatic detection and correction of non-red eye flash defects | |
| CN109472271B (zh) | 印刷电路板图像轮廓提取方法及装置 | |
| CN105976354B (zh) | 基于颜色和梯度的元件定位方法和系统 | |
| CN112686858A (zh) | 一种手机充电器视觉缺陷检测方法、装置、介质及设备 | |
| WO2017071406A1 (zh) | 金针类元件的引脚检测方法和系统 | |
| WO2018068414A1 (zh) | 色环电阻的检测方法、装置和自动光学检测系统 | |
| CN108918093B (zh) | 一种滤光片镜面缺陷检测方法、装置及终端设备 | |
| CN115205223B (zh) | 透明物体的视觉检测方法、装置、计算机设备及介质 | |
| WO2017107529A1 (zh) | 一种并排二极管的定位方法及装置 | |
| WO2018010386A1 (zh) | 元件反件检测方法和系统 | |
| US20210325299A1 (en) | Digital assessment of chemical dip tests | |
| TWI512284B (zh) | 玻璃氣泡瑕疵檢測系統 | |
| CN113034488A (zh) | 一种喷墨印刷品的视觉检测方法 | |
| KR102242996B1 (ko) | 자동차 사출제품의 비정형 불량 검출 방법 | |
| CN116168218A (zh) | 一种基于图像识别技术的电路板故障诊断方法 | |
| CN115619725A (zh) | 电子元器件检测方法、装置、电子设备及自动质检设备 | |
| JP2017142219A (ja) | 画像検査装置、画像検査プログラム及びコンピュータで読み取り可能な記録媒体並びに記録した機器 | |
| TWI510776B (zh) | 玻璃氣泡瑕疵檢測處理方法 | |
| CN110782453B (zh) | 一种基于图像特征匹配的印刷品质量检测方法 | |
| CN114354491A (zh) | 一种基于机器视觉的dcb陶瓷基板缺陷检测方法 | |
| CN112164058A (zh) | 用于滤光片的丝印区域粗定位方法、装置及存储介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16918909 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 04/10/2019) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16918909 Country of ref document: EP Kind code of ref document: A1 |