KR101358270B1 - Method and apparatus for controlling humanoid robot arm - Google Patents

Abstract

The present invention discloses a humanoid robot arm control apparatus and control method. The robot arm control apparatus according to the present invention includes an imaging unit for imaging an image, an allocator for allocating a level value according to the intensity level of each pixel based on pixel information of the captured image, and a contrast to the corresponding pixel based on the level value. An extraction unit for extracting the trajectory data of the robot arm for drawing a line and a control unit for controlling the robot arm to draw an image according to the trajectory data.

Description

Humanoid robot arm control method and apparatus {METHOD AND APPARATUS FOR CONTROLLING HUMANOID ROBOT ARM}

The present invention relates to a method and apparatus for controlling a humanoid robot arm, and more particularly, to a method and apparatus for controlling a humanoid robot arm to express contrast when drawing an image by a humanoid robot.

Humanoid robots are robots whose behavior, appearance and artificial intelligence (AI) system are aimed at human beings. In particular, the humanoid painter who draws through a robot arm must move like a human in the manner or order of drawing, and the movement trajectory of the robot arm must be optimized to save time and improve the completeness of the painting. In the case of the existing painter robot, in general, the outline and contour of the image are inputted, and the trajectory for controlling the actual robot arm is extracted from the processed image, and only the image is drawn.

Recently, as the robot industry is rapidly growing, the robotic art field of the professional service robot field is rapidly growing, and researches on humanoid robots drawing through robot arm control are actively conducted. Techniques such as Canny Edge Detector and S. Calinon's Haar-like feature were used to recognize specific objects, extract the outlines and outlines of recognized object images, and draw a robot. In addition, C. Lin et al. Have already studied portraits using humanoid robot Pica. And Chinese Painting Robot, produced by F. Yao and G.Shao, can draw ink paintings using real brushes as drawing materials. However, it is pointed out that the conventional method cannot draw shadows or contrasts of an image by drawing only the outline and the outline of the image and drawing the image.

The present invention has been made to solve the above problems, an object of the present invention, the humanoid robot arm control method for controlling the robot arm to distinguish the contrast of the input image and express the contrast in the density of the contrast line And to provide an apparatus.

In order to achieve the above object, the robot arm control apparatus according to the present invention includes an image capturing unit for capturing an image, and assigning a level value according to the intensity level of each pixel based on pixel information of the captured image. An allocator, an extractor which extracts trajectory data of a robot arm for drawing a contrast line on a corresponding pixel based on the level value, and a controller that controls the robot arm to draw the image according to the trajectory data.

In this case, the allocator compares the brightness values of the pixels constituting the image with a plurality of predetermined range values, and corresponds to the range values when included in any one of the plurality of range values. The level value may be assigned to a value that is obtained.

The extraction unit may include an operation unit that calculates the number of repetitions of the contrast line passing through the pixel based on a level value, and an alignment unit that determines the order of the direction in which the contrast line is drawn according to the number of repetitions.

In this case, the contrast line, the robot arm control device, characterized in that for removing a portion of the contrast line according to the following equation:

는 명암 레벨 값의 집합이고, S_w는 이미지의 폭, S_h는 이미지의 높이이다.here,

Is the set of contrast level values, S _w is the width of the image, and S _h is the height of the image.

According to another aspect of the present invention, there is provided a method for controlling a robot arm, the method comprising: imaging an image, allocating a level value according to a brightness level of each pixel based on pixel information of the photographed image, and based on the level value And extracting trajectory data of the robot arm for drawing a contrast line on the corresponding pixel, and controlling the robot arm to draw the image according to the trajectory data.

In this case, the step of allocating the level value may include comparing the brightness value of the pixels constituting the image with a plurality of predetermined range values, and when the level value is included in any one of the plurality of range values, Each level can be assigned a level value.

The extracting of the trajectory data may include calculating a repetition number of the contrast line passing through the pixel based on a level value and determining an order of the direction in which the contrast line is drawn according to the repetition number. can do.

In this case, the contrast line, the robot arm control method characterized in that a part of the contrast line is removed according to the following equation:

는 명암 레벨 값의 집합, S_w는 이미지의 폭, S_h는 이미지의 높이이다.here,

Is the set of contrast level values, S _w is the width of the image, and S _h is the height of the image.

According to various embodiments of the present disclosure as described above, the trace data corresponding to the robot arm is extracted to map not only the shape of the image but also the contrast such as the shadow, and the robot maps the map according to the size of the actual object to be drawn in real time. The robot arm can be controlled to express an image.

1 is a view for explaining the leveling of the contrast density of an image according to an embodiment of the present invention,
2 is a block diagram illustrating a robot arm control apparatus according to an embodiment of the present invention;
3 is a view for explaining a contrast expression according to an embodiment of the present invention;
4 is a view for explaining the contrast step image according to an embodiment of the present invention,
5 is a view for explaining an image according to the number of contrast lines according to an embodiment of the present invention;
6 is a flowchart illustrating a robot arm control flow according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention.

1 is a view for explaining the intensity leveling of the image according to an embodiment of the present invention.

When a color image is input through a camcorder or other video device, the intensity level is leveled to extract the trajectory of the robot's movement from the input image.

When the input image is in color mode or black and white mode, that is, when pixel values of each image are defined by RGB values, the color image is gray scale because it is difficult to extract the robot's trajectory with these pixel values. Is converted to.

For each pixel constituting the converted gray scale image, each pixel is leveled according to the intensity density. That is, the pixels constituting the image have a value between 0 and 255 depending on the brightness. When the intensity level is expressed in five levels, the intensity level may be defined as any one of five levels according to the brightness value of each pixel.

Specifically, when the pixel value has a value between 1 and 51, the contrast density level of the pixel is defined as the first level. If the pixel value has a value between 51 and 102, the intensity density level of that pixel is defined as the second level. In this manner, the intensity density leveling is performed on all pixels constituting the gray scale image. The larger the level value of the pixel on which the intensity density leveling is performed, the darker the portion, and the smaller, the brighter the portion.

Referring to FIG. 1, it can be seen that the level values of the image on which the contrast density leveling is performed are displayed in an array. As illustrated in FIG. 1, a pixel having a level value of 0 means a pixel of the brightest area, and a pixel of a portion having a level value of 4 means a pixel of the darkest area.

In FIG. 1, pixels constituting the image on which the intensity density leveling is performed may be represented by a set of level values of a two-dimensional array. That is, the index of each array represents the coordinate value of the image. One-dimensional and two-dimensional size of this arrangement means the size of the horizontal and vertical of the image.

From this level value, contrast lines such as the first straight line (horizontal arrow) and the second straight line (diagonal arrow) shown in FIG. 1 may be extracted to express the contrast of the image. First, the direction is defined by varying the direction of the contrast line for each level. For example, as illustrated in FIG. 1, the first stage has a directional direction in the horizontal direction, and the second stage has a direction in the vertical direction. From the next stage, the contrast line is expressed by adjusting the direction angle value according to each stage. There are two main factors that express the contrast.

The first element is the overlay of the contrast line. If contrast is expressed at five levels, level 5 has the brightest intensity and is not represented with contrast lines. In case of level 4 and below, draw a contrast line with level 4 direction. Thereafter, the level 3 or less is drawn over the level 4 contrast line with the contrast line of the level 3 level. When the contrast line is drawn in this way, the contrast line with a total of four directional lines is drawn in the first stage, and the density of the line is increased to express the darkest contrast. From the next level, the overlap of lines gradually decreases to express low contrast.

The second element is the number of contrast lines. Increasing the number of contrast lines in each level has a more dense contrast line, which increases the completeness of the picture. However, the drawing speed is slower because the robot arm has more trajectories to draw. Simply extracting the contrast lines from the leveled coordinates can lead to short or dark lines that are not valid when drawn by a real robot. Alternatively, there may be a portion where the robot arm takes too much time to draw, even if it is a valid contrast line. Thus, invalid short and dark lines can be eliminated by Equation 1 below.

Where S _w is the width of the image, S _h is the height of the image, and L _shade is the set of pixel intensity levels. L _crrent represents the level value of the current coordinate, and term is a variable that determines the validity of the contrast line. That is, a scan for each line and level, and inspection of the line is less than the sum of the total level value of the term as long as the right by the term to the left in the L _current operation and then the average value, and that the calculated value term value If the average value is greater than the term value, it is determined as an invalid level value.

Through this process, depending on the term's setting, the dark and dark lines can be removed.

2 is a block diagram for explaining a robot arm control apparatus according to an embodiment of the present invention.

Referring to FIG. 2, the robot arm control apparatus according to the present invention includes an imaging unit 210, an allocation unit 230, an extraction unit 250, and a control unit 270.

The imaging unit 210 may be implemented as a camcorder or other video device. The imaging unit 210 may be a scan type input device capable of receiving an external image, or may be implemented as a digital camera. Alternatively, the imaging unit 210 may be implemented as another type of input device capable of receiving an image.

The allocator 230 receives an image captured by the imager 210, analyzes pixel information of the image, and allocates a level value according to the intensity level of each pixel based on the pixel information of the image. That is, the allocator 230 compares brightness values of pixels constituting the image with a plurality of predetermined range values. In this case, the plurality of range values mean a range partitioned by uniformly dividing the brightness value of the pixel, a value between 0 and 255, to a predetermined size. If a brightness value of a specific pixel belongs to any one of a plurality of range values, the level value of the corresponding pixel may be assigned to a value corresponding to a range value to which the brightness value of the corresponding pixel belongs. More details will be described later.

The extraction unit 250 extracts the trajectory data of the robot arm for drawing a contrast line on the pixel based on the level value. That is, the contrast line is repeatedly drawn on the pixel as many times as the level value, and the repeated contrast line is sequentially drawn in the horizontal direction, the vertical direction, the first diagonal direction, and the second diagonal direction. .

The extraction unit 250 extracts trajectory data including information on the number of times the contrast line overlaps with each pixel of the image and information about the relationship between the contrast lines between the pixels. Such trajectory data may be mapped such that the robot arm fits the size of the object to which the image is drawn.

In this case, the extractor 250 calculates the number of repetitions of the contrast line passing through the corresponding pixel based on the level value and the horizontal direction, the vertical direction, the first diagonal direction, and the second direction, according to the number of repetitions. It may include an alignment unit (not shown) for determining the order in which the contrast line is sequentially drawn in the diagonal direction.

The controller 270 may control overall operations of the imaging unit 210, the allocation unit 230, and the extraction unit 250, and control the robot arm to draw an image according to the trajectory data.

3 is a view for explaining the contrast expression step according to an embodiment of the present invention.

Referring to FIG. 3, FIG. 3A illustrates leveling coordinates in which light and dark concentrations are divided in five steps. As shown in FIG. 3 (a), the brightest part does not display the contrast line. It can be seen that four contrast lines are drawn according to the level leveled according to the intensity level.

In addition, referring to FIG. 3 (b), the leveling coordinates shown in FIG. 3 (a) are drawn with the same contrast lines. At this time, the contrast line is drawn only for the remaining pixels (levels 1 to 4) except for the brightest pixel (level 5) (first step).

In addition, referring to FIG. 3 (c), in the first stage contrast expression step shown in FIG. 3 b, the contrast line for the second brightest pixel (level 4) is removed, and the remaining level values (levels 1 to 3). Draw a contrast line only for pixels of ().

In addition, referring to FIG. 3 (d), the contrast line for the third brightest pixel (level 3) is removed in the second stage contrast expression step shown in FIG. 3 (c), and the remaining level values (levels 1 and 2). Draw a contrast line only for pixels of () (step 3).

Referring to FIG. 3 (e), in the third level contrast expression step shown in FIG. 3d, the contrast line for the fourth brightest pixel (level 2) is removed, and the remaining level value (level 1) is removed. Draw a contrast line only for pixels (step 4).

As described above, the input image may have a large difference depending on the intensity level of the contrast and the number of contrast lines. Next, we will compare the original image with the appropriate number of contrast lines and density levels.

4 (a) to 4 (d) are diagrams for explaining an image of contrast steps according to an embodiment of the present invention.

4 (a) shows the original image. The original image has not been leveled by contrast density. The image leveled at the 3, 4, and 5 levels of the original image is compared with the original image.

First, FIG. 4B shows an image of leveling the original image of FIG. 3A with three levels of contrast. Referring to FIG. 4 (b), since the image of FIG. 4 (b) is leveled with the original image by three levels, when compared with the original image, pixels of the third bright area are removed and only pixels of darker areas are represented. You can see that the contrast is rougher than the original image.

4 (c) shows an image of leveling the original image of FIG. 4 (a) with four levels of contrast levels. Referring to FIG. 4 (c), since the image of FIG. 4 (c) is leveled at the original image by 4 levels, when compared with the original image, the pixel of the fourth bright region is removed and only the pixels of the darker region are represented. Although the contrast is rougher than the original image, it can be seen that the contrast is finer than the image of FIG.

FIG. 4 (d) shows an image of leveling the original image of FIG. 4 (a) with five levels of contrast. Referring to FIG. 4 (d), since the image of FIG. 4 (d) is leveled at the original image by 5 levels, when compared with the original image, only the brightest pixel is removed and the darker pixel is represented. Although the contrast is somewhat rougher than the original image, it can be seen that the contrast is more detailed than the image of FIGS. 4 (b) and 4 (c).

5 (a) to 5 (d) are views for explaining an image according to the number of contrast lines according to an embodiment of the present invention.

FIG. 5 (a) shows an image in which 20 contrast lines represent the original image shown in FIG. 4 (a). Referring to the drawing of FIG. 5 (a), when the number of contrast lines is 20 compared to the original image, it can be confirmed that the outline of the object is somewhat rough and not clearly expressed.

FIG. 5 (b) shows an image in which 30 contrast lines are used to represent the original image shown in FIG. 4 (a). As shown in FIG. 5 (b), when the original image is represented by 30 contrast lines, the outline and contrast expressions of the whole object are considerably lower than those of the original image, but compared to the image shown in FIG. 5 (a). It can be seen that it is expressed in more detail.

FIG. 5 (c) shows an image in which the original image shown in FIG. 4 (a) is represented by 40 contrast lines. As shown in FIG. 5 (c), when the original image is represented by 40 contrast lines, the outline and contrast expressions of the whole object are somewhat inferior to the original image, but compared to FIGS. 5 (a) and 5 (b). We can see that the contrast is fine.

FIG. 5 (d) shows an image in which 50 contrast lines represent the original image shown in FIG. 4 (a). As shown in (d) of FIG. 5, when the original image is represented by the number of 50 eye lines, the outline and contrast expressions of the whole object are slightly lower than those of the original image, but FIGS. 5 (a), 5 (b) and Compared with 5 (c), the contrast is finer and sufficient to distinguish the contrast of the original image.

6 is a flow chart for explaining the flow of the robot arm control method according to an embodiment of the present invention.

Referring to FIG. 6, the robot arm control method according to the present invention includes the steps of image capturing (S610), assigning a level value (S630), extracting trajectory data (S650), and controlling a robot arm ( S670).

In step S610, the image is captured and an image to be drawn is obtained to obtain pixel information values of the image.

In operation S630, the level value is assigned according to the intensity level of each pixel based on the pixel information of the picked-up image. The contrast level may be determined as a plurality of levels, but it is preferable to level the contrast level in five levels. At this time, the step of assigning the level value (S630), when the brightness value of the pixels constituting the image is included in any one of the plurality of range values, and corresponding to the corresponding range value, respectively Assign a level value by value.

Extracting the locus data (S650) extracts the locus data of the robot arm for drawing a contrast line on the pixel based on the level value assigned to each pixel. At this time, the trajectory data representing the contrast can be extracted by adjusting the density of the contrast line according to the pixel level value. The density of the contrast line extracts the trajectory data so that the density can be adjusted by repeatedly overlapping the contrast line at a specific pixel. In addition, the trajectory data includes information about a connection relationship between light and dark lines between pixels, and also includes information about the number of light and dark lines overlapping with a specific pixel.

In the extracting of the locus data (S650), the number of repetitions of the contrast line passing through the corresponding pixel is calculated based on the level value, and according to the number of repetitions, the horizontal direction, the vertical direction, the first diagonal direction, and the second The trajectory data including the order in which the contrast lines are drawn sequentially in the diagonal direction is extracted.

In the controlling of the robot arm (S670), the robot arm is controlled to draw the captured image on the object to be drawn according to the trajectory data.

Although illustrative embodiments and applications of the present invention have been shown and described, many changes and modifications may be made without departing from the scope of the present invention, and such modifications may be made by one of ordinary skill in the art to which the present invention pertains It can be clearly understood. Accordingly, the described embodiments are illustrative and not restrictive, and the invention is not limited by the accompanying detailed description, but is capable of modifications within the scope of the claims.

200: robot arm control device 210: imaging unit
230: allocation unit 250: extraction unit
270:

Claims (8)

An imaging unit for imaging an image;
An allocator configured to allocate a level value according to the intensity level of each pixel based on pixel information of the photographed image;
An extraction unit for extracting trajectory data of a robot arm for drawing a contrast line on a corresponding pixel based on the level value; And
And a controller configured to control the robot arm to draw the image by expressing contrast according to the trajectory data.
Wherein the allocating unit comprises:
The brightness value of the pixels constituting the image is compared with a plurality of predetermined range values, and when the range value is included in any one of the plurality of range values, the level value is allocated to a value corresponding to each of the range values. Robot arm control device, characterized in that. delete The method of claim 1,
The extracting unit extracts,
A calculator configured to calculate the number of repetitions of the contrast line passing through the corresponding pixel based on a level value; And
And an alignment unit for determining an order of the direction in which the contrast line is drawn according to the number of repetitions. The method of claim 3,
The contrast line is a robotic arm control device, characterized in that for removing a portion of the contrast line according to the following equation:

here,

Is the set of contrast level values, S _w is the width of the image, and S _h is the height of the image. Imaging an image;
Assigning a level value according to the intensity level of each pixel based on pixel information of the photographed image;
Extracting trajectory data of a robot arm for drawing a contrast line on a corresponding pixel based on the level value; And
And controlling the robotic arm to draw the image by expressing contrast according to the trajectory data.
Assigning the level value,
The brightness value of the pixels constituting the image is compared with a plurality of predetermined range values, and when the range value is included in any one of the plurality of range values, the level value is allocated to a value corresponding to each of the range values. Robot arm control method characterized in that. delete 6. The method of claim 5,
Extracting the trajectory data may include:
Calculating the number of repetitions of the contrast line passing through the corresponding pixel based on a level value; And
And determining the order of the direction in which the contrast line is drawn according to the number of repetitions. The method of claim 7, wherein
The contrast line is a robot arm control method characterized in that a part of the contrast line is removed according to the following equation:

here,

Is the set of contrast level values, S _w is the width of the image, and S _h is the height of the image.

Images