JPS59188784A - Method for recognizing illuminant - Google Patents

Abstract

PURPOSE:To obtain accurately the number of illuminant by adjusting the threshold value so that the difference in a level meter due to fluctuation of lightness is eliminated from the information of two areas of a continuous illuminants thereby finding out the regularity of the continuous illuminants. CONSTITUTION:The level meter in an area 2 is recognized at first, the width 6 of the level meter is checked from left to right of the area, the change in the width and the variation in the position are checked so as to obtain the range of the size of the level meter. Then, the part within this range is checked lengthwise, the threshold value is increased until the illuminant separated at an interval having the regularity is obtained and when the illuminant separated by a prescribed interval is obtained, the length for one illuminant, the average value of the interval of the illuminants and the number are obtained, the length of the lighting part is obtained from the number, the length of the illuminant and the average value of the interval, the relation with the length of the lighted part of the level meter obtained at first is checked so as to decide the number of the illuminants.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention uses a visual recognition system, in particular a light emitter (LED).
The present invention relates to a method for recognizing a light emitter suitable for recognizing the lighting amount of a level meter in an audio system, etc.).

[Prior art]

In conventional recognition, the target was a reflective object rather than a light-emitting body, so information based on area could be used to adjust the threshold. In this case, it is necessary to separate each of the light emitting bodies and find the number of lights on in the light, and there is a drawback that it cannot handle cases where the number of lights on one point changes.

[Purpose of the invention]

The object of the present invention is to use a visual recognition processing system to
The number of lit light emitters in a level meter such as an audio system using LEDs, etc. can be recognized without being affected by the interval between the light emitters of the level meter or fluctuations in brightness within the level meter. It is about providing.

[Summary of the invention]

To achieve the above object, the present invention provides an area for obtaining image information in a part of the continuum in order to accurately determine the edges of the continuum when recognizing continuous illuminants. Accurately determine the edge in the width direction of the body by adjusting the threshold,
Further, a sensor is installed in a different location, and the threshold value is adjusted to eliminate the difference due to brightness fluctuations in the level meter from the information of the two areas, and the size of the continuum is determined. Furthermore, the method is characterized in that the threshold value is adjusted to find regularity in the continuum, and from the contents of the regularity, the number of lit light emitters is determined.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to FIGS. 1 to 4.

The configuration required to carry out the present invention consists of a television camera, a monitor, and an image processing device that can perform image processing with binary images, and can set an area at the recognition position of the level meter by inputting markers or numerical values. shall be.

FIG. 1 shows the appearance of the level meter.

As shown in the figure, a level meter 1 is one in which items of the same size are lined up in a row at the same intervals.

As a recognition area for obtaining image information of the lit portion of the level meter 1, an area 2 in the figure is provided in a part of the level meter slightly larger than the level meter.

The level meter 1 is photographed with a television camera, taken into an image processing device, and subjected to binarization processing.

2 to 4 show a binarized image 4 of the level meter when it is binarized.

FIG. 2 shows the optimal value of the threshold, FIG. 3 shows the state when the threshold is low, and FIG. 4 shows the state when the threshold is high.

First, the level meter in area 2 is recognized, and the width 6 of the level meter is examined from the five sets of areas to the right, and changes in the width and position (up and down in the figure) at that time are examined.

Compare the average width 6 and the width 7 of the area to see if the average width is the width 7 of the area.

If not, raise the threshold.

If the average width of the level meter is 6 and the widths of areas 7 and 7z are two, adjust the threshold so that the fluctuation in the width direction of the width of area 7 is reduced. Area 3 is provided as a recognition area.

Area 2. The average width of the level meters in each area 3 is checked, and if the average difference between both widths is small, the areas of the lighting parts in both areas are compared. When the difference in the average width is large, the area 6 is moved to find a point where the difference in the average width is small and the areas are compared.

The areas in both areas are compared, and if the difference between the areas is large, the threshold value is adjusted while the width within area 2 does not change to reduce the difference between the areas.

If the difference between areas is small, area 2. Move in both directions in area 6 and detect both ends of the level meter.

At this time, the average value of the widths in both areas of the level meter is taken as the width of the level meter, and the distance between both ends is taken as the length of the level meter, and the size of the entire lighting part of the level meter is determined. (Fig. 5) Examine the range of the determined level meter size in the length direction, and check the distribution of the lighting part within the level meter.
It is determined whether the light emitting bodies are one light emitting body or the light emitting bodies are separated at regular intervals.

In the case of one continuous light emitting body, the threshold value is increased to find a point where the light emitting bodies are separated at regular regular intervals.

(Situation between Figure 4 and Figure 2) When the luminous bodies are separated at regular intervals, the length and spacing of the luminous bodies are averaged, and the luminous bodies 1
Determine the length of the minutes, the average distance between the light emitters, and the number of light emitters,
From the number of light emitters and the average length/interval of one light emitting body, the length of the entire lighting section is determined, and the relationship between the length component and the length of the entire lighting section of the level meter, which was initially determined, is determined. If there is no difference in length between the two, the number of light emitters at that time is taken as the number of lit light emitters in the level meter. Furthermore, if there is a difference in length between the two, the threshold value is further adjusted to find a point with no difference in length.

According to this embodiment, the number of lights on the level meter can be determined without being affected by the difference in the number of light emitters in the level meter - the size and spacing of the light emitters, or the difference in brightness of the light emitters in the level meter. Therefore, it is effective as a new recognition function because it can recognize the number of lights on regardless of the level meter.

〔Effect of the invention〕

According to the present invention, in a recognition processing system that performs visual recognition processing using a binary image, recognition of the lighting amount of consecutive light emitters is performed based on the size of the light emitters, the difference in the interval between the light emitters, and the brightness of the light emitters. The size of the continuum can be found without being affected by
Furthermore, since the number of luminous bodies in the continuum can be accurately determined by separating the luminous bodies in the continuum into units of luminous bodies, this method is effective as a new method for recognizing luminous bodies in visual recognition.

[Brief explanation of drawings]

Figure 1 is an external view of the level meter, Figure 2 is the optimal threshold value when the level meter in Figure 1 is converted into a binary value, and Figure 6 is the threshold value when the level meter in Figure 1 is converted into a binary value. When is high, Figure 4 is the level meter in Figure 1. When the threshold value is low when binarized, Figure 5 is a diagram where both ends of the level meter are found and the size is determined. 1... Appearance of the level meter 2... Recognition area 6... Recognition area 4...
Binarized level meter 5... Components at the time of binarization 6... Level meter width 7... Area width 8...
Length of level meter No. 1 Figure 2 Figure 5 Figure 4 Figure 5

Claims (1)

[Claims] In order to recognize and determine the number of successive small light emitters turned on in a visual recognition system using binarized images, two areas are used as recognition areas to obtain image information of the light emitters at the positions of successive light emitters. The device is equipped with a function that allows the threshold value to be freely adjusted, and has the ability to make judgments for recognition. A luminous object recognition method characterized by providing a function for determining the .