KR20100064634A - Apparatus for detecting and eliminating insect in digital image processing device - Google Patents

Abstract

The present invention relates to an apparatus for detecting and eradicating bugs in a digital image processor which can solve a cause of an error and a malfunction caused by the insect by detecting and erasing a bug with a digital image processor. In the digital image processor, the worm detection and removal apparatus includes worm detection means for detecting presence or absence of a worm of the target from feature extraction information of the target and / or motion information of the target; And a worm control means for combating the detected worm by using a mechanism provided therein when the worm detection signal is received from the worm detection means.

Description

Apparatus for detecting and eliminating insect in digital image processing device}

The present invention relates to a device for detecting and combating bugs in a digital image processor.

If the worm is located near the digital image processor, it will focus on the worm, not AF, AE, or AWB, instead of the AF, AE, or AWB that was originally targeted by the bug. AF, AE, and AWB will be activated, making the entire background of the screen, including the monitored object, blurred and difficult to see. If the worm is located remotely from the digital image processor, it will cause errors in missing or missing objects. In addition, if the insect attack or the body is caught between the parts of the digital image processor, it may cause a breakdown or damage, and the life of the digital image processor is shortened.

The technical problem to be solved by the present invention is to provide a bug detection and eradication device in the digital image processor that can solve the cause of the operation errors and failures caused by the bug by detecting and combating the bug with the digital image processor. .

In the digital image processor for solving the technical problem to be solved by the present invention, the bug detection and eradication apparatus is a digital image processor, which detects the presence or absence of the insect bug from the feature extraction information and / or motion information of the target Worm detection means; And when the worm detection signal is received from the worm detection means, it is preferable to include a worm control means for combating the detected worms using a mechanism provided therein.

In the present invention, the worm detection means may detect the presence or absence of worms from the voice information of the target or the moving vibration frequency of the target.

In the present invention, the worm detection means includes a database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A scan unit for setting a predetermined area and scanning the area from an input image; An adjusting unit for adjusting the size of the input image; And a search unit for searching for a worm by applying the determination criteria of each scanned image whose size is adjusted.

In the present invention, the worm detection means comprises a comparison unit for comparing the previous frame and the current frame; A generator configured to generate a moving object candidate group according to the comparison result; And a detector configured to detect the moving object having a predetermined size or less from the candidate group.

In the present invention, the bug detecting means comprises: an estimator for estimating a background from previous frames; A comparison unit comparing the estimated background with a current frame; A generator configured to generate a moving object candidate group according to the comparison result; And a detector configured to detect a moving object having a predetermined size or less from the candidate group.

In the present invention, the worm detection means includes a first comparison unit for comparing the previous frame and the current frame; A first generator configured to generate a moving object candidate group according to the first comparison result; A first detector configured to detect the moving object having a predetermined size or less from the first candidate group; A second comparing unit comparing the current frame with the next frame; A second generation unit which generates a moving object candidate group according to the second comparison result; A second detector detecting the moving object having a predetermined size or less from the second candidate group; A third comparing unit comparing a position of the first detected object and a position of the second detected object; And a determination unit determining that there is a movement at the closest position as a result of the third comparison.

In the present invention, the worm detection means includes an analysis unit for analyzing the shape of the movement from the previous frame; A selecting unit which selects a position candidate group in which a motion is expected in a current frame from the analyzed motion type; A comparator for comparing the shapes of the candidate group and the moving sieve; And as a result of the comparison, it may include a determination unit that determines to move to the most similar candidate group.

In the present invention, the first analysis unit for determining the irregularity of the movement direction by analyzing the motion trajectory of the object; A second analyzing unit which determines whether the moving direction is variable by analyzing a speed of an irregular object; And it may further comprise a determination unit for determining that the movement direction is irregular, the variable speed is an insect.

In the present invention, the worm detection means includes a first comparison unit for comparing the previous frame and the current frame; A generator configured to generate a moving object candidate group according to the first comparison result; A first detector detecting the moving object having a predetermined size or less from the candidate group; A database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A second comparing unit comparing a shape of the detected moving object with a shape of the determination criterion; And a second detection unit detecting a worm as a result of the second comparison.

In the present invention, the worm control means includes an estimator for estimating a background from previous frames; A first comparing unit comparing the estimated background with a current frame; A generator configured to generate a moving object candidate group according to the first comparison result; A first detector detecting a moving object having a predetermined size or less from the candidate group; A database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A second comparison unit comparing a shape of the first detected moving object with a shape of the determination reference; And a second detection unit detecting a worm as a result of the second comparison.

In the present invention, the worm control means is a worm detected by including a flapper, a chemical injector, a smoke generator, a light or ultraviolet emitter, a tattoo sticker for combating insects, a sound wave or an ultrasonic wave generator, an impact bomber, an insect attractor, etc. Can fight.

As described above, according to the present invention, by detecting and erasing a bug with a digital image processor, it is possible to solve the causes of operational errors and failures caused by the bug.

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

1 is a block diagram showing the configuration of a bug detecting and erasing device in a digital image processor according to the present invention, which includes a bug detecting means 100 and a bug erasing means 200.

Bug detection means 100 detects bugs using video or audio information. However, since it is impossible to ensure the performance of 100 percent detection, it also includes the case of direct detection (manual detection) by the human eye in some cases.

The insect exterminating means 200 operates by the insect detecting signal of the insect detecting means 100, and operates the bed restraint mechanism provided in the digital image processor to combat the insect.

Considering that there are many other insects in the evening, it is also possible to install a timer (not shown) so that the insect control means 200 automatically operates at the evening. In addition, it is also possible to install a timer in consideration of the case where the worm is installed in a large number of times, so that the worm control means 200 operates automatically at specific time intervals.

The bug detecting means 200 may detect bugs by using image processing and voice processing.

In the case of larvae species, the method of image processing is different. After learning the shape, we extract the 8-legged feature and search the input video. In the case of worm species, after learning the shape of the worm, it extracts features such as wings and searches for images, or detects and traces moving objects based on the irregular and fast movement of the worm, and then analyzes the trajectory of the bee range. Alternatively, a moving range can be detected by comparing the shape of a blade with a moving object.

There are two ways to use voice processing: the sound of insects and the vibration frequency of the insect's wings. In the case of using the sound of the bug, the sound around the digital image processor is recorded, the background noise is removed, and the existence of the bug is determined by comparing the similarity with the sound of the bug. In order to detect the vibration frequency of the insect's wing movement, the vibration frequency range can be set through the movement data of the insect wing, the vibration around the digital image processor is detected, the noise is removed, and the vibration frequency range is confirmed and compared. Determine the presence of bugs. At this time, the wing of the worm has a very fast movement and has a vibration frequency range of the high frequency band.

In the case of a method using voice processing, detection is possible only when the bug is close to the digital image processor. On the other hand, in the case of the method using the image processing, if the bug is attached to the digital image processor or located in a square, the bug cannot be detected, and thus the performance can be improved by complementary relations when detecting the bug.

Next, the insect detecting means 100 will be described in detail with reference to FIGS. 2 to 10.

FIG. 2 is a detailed block diagram of the worm detection means 200 according to the first embodiment of FIG. 1, and may be applied to worm detection regardless of larvae, spiders, or insects. The bug detecting means 200 of FIG. 2 includes a database 121, a learning unit 122, a determination criterion generating unit 123, a scanning unit 124, a size adjusting unit 125, and a searching unit 126. The insect detecting means 200 shown in FIG. 2 is a method of detecting an entire insect by searching for a bug shape and then searching the entire input image to see if there is anything that matches the shape.

10 illustrates a method used for face detection of a person. A criterion for discriminating the presence or absence of a human face is generated by learning the images of the human face and the non-face image with information such as brightness difference and edge of a specific area in a database. Afterwards, the face of the person is found by applying the created criterion while adjusting the size of the entire image. However, in the present invention, the shape of a worm is used instead of the shape of a human.

As shown in FIG. 10A, the database 121 stores an image which is not a worm shape and an image which is a worm shape. As shown in FIG. 10B, the learner 122 learns the presence or absence of a worm by characterizing information such as brightness difference and edge of a specific region in the database 121. The determination criteria generation unit 123 generates a determination criterion that can distinguish the presence or absence of bugs as a learning result. The scan unit 124 scans the input image with a predetermined area set in advance as shown in FIG. 10C. In this case, the size adjusting unit 125 adjusts the size of the input image, and the scanning unit 124 scans the area while adjusting the size of the input image. The search unit 126 searches for bugs by applying the criteria generated above to the scan result.

3 is a detailed block diagram of the insect detecting means 200 according to the second embodiment of FIG. 1, which detects insects by using movement, and is used when detecting insect species. The bug detecting means 200 of FIG. 3 includes an image comparator 131, a posterior group generator 132, and a motion detector 133.

The image comparator 131 compares the image of the previous frame with the image of the current frame. The candidate group generation unit 132 generates a candidate group of moving objects with respect to the part having a difference as a result of the image comparison. The motion detector 133 detects a moving object having a predetermined size or less from the generated candidate group. This means that moving objects smaller than a certain size are recognized as insects. Hereinafter, referring to FIG. 7, the contents of detecting a bug from the moving object will be described.

FIG. 4 is a detailed block diagram of the insect detecting means 200 according to the third embodiment of FIG. 1, which detects insects by using movement, and is used when detecting insect species. The bug detecting means 200 of FIG. 4 includes a background estimator 141, an image comparator 142, a posterior group generator 143, and a motion detector 144.

The background estimator 141 estimates the background of the object currently being photographed from the plurality of previous frames. The image comparator 142 compares the current frame image with the estimated background image. In other words, the background image is subtracted from the current frame image to extract an image having a difference. The candidate group generation unit 143 generates the image having the difference as the candidate group of the moving object based on the comparison result of the current frame image and the estimated background image. The motion detector 144 detects a moving object having a predetermined size or less from the generated candidate group. This means that moving objects smaller than a certain size are recognized as insects. Hereinafter, referring to FIG. 7, the contents of detecting a bug from the moving object will be described.

FIG. 5 is a detailed block diagram of the insect detecting means 200 according to the fourth embodiment of FIG. 1, which detects insects by using movement, and is used when detecting insect species. The bug detecting means 200 of FIG. 5 includes a first image comparator 151, a first candidate group generator 152, a first motion detector 153, a second image comparator 154, and a second candidate group generator. 155, a second motion detector 156, a position comparator 157, and a motion determiner 158.

The first image comparator 151 compares the image of the previous frame with the image of the current frame. The second candidate group generation unit 152 generates a candidate group of moving objects with respect to a portion that is different as a result of the first image comparison. The first motion detector 153 detects a moving object having a predetermined size or less from the generated first candidate group. The second image comparator 154 compares the image of the current frame with the image of the next frame. The second candidate group generation unit 155 generates a candidate group of moving objects with respect to a portion that is different from the second image comparison result. The second motion detector 156 detects a moving object having a predetermined size or less from the generated second candidate group. The position comparison unit 157 compares positions of the first detected moving object and the second detected moving object. As a result of the position comparison, the motion determining unit 158 determines that there is motion with respect to the object when the first detected moving object and the second detected moving object are in the closest position. Hereinafter, referring to FIG. 7, the contents of detecting a bug from the moving object will be described.

FIG. 6 is a detailed block diagram of the bug detecting means 200 according to the fifth embodiment of FIG. 1, which detects insects by using movement, and is used when detecting insect species. The bug detecting means 200 of FIG. 6 includes a motion shape analyzer 161, a candidate group selector 162, a shape comparison unit 163, and a motion determiner 164.

The motion shape estimator 161 analyzes the motion shape from the plurality of previous frames. The candidate group generation unit 162 selects a position candidate group in which a motion is expected in the current frame by applying a motion pattern analyzed for the current frame. The shape comparison unit 163 compares the shape of the moving object in the current frame with the shape of the selected candidate group. As a result of the comparison, the motion determination unit 164 determines that the object moves to the candidate group having the highest similarity in the current frame. Hereinafter, referring to FIG. 7, the contents of detecting a bug from the moving object will be described.

FIG. 7 is a block diagram showing a configuration of determining a bug according to the motion detection in FIGS. 3 to 6, and includes a motion trajectory analyzer 171, a speed analyzer 172, and a bug determiner 173.

The motion trajectory analyzer 171 tracks the detected or determined moving object and analyzes the trajectory of the motion, and then determines whether the movement direction is regular or irregular. This is possible because the movement of the beetle is irregular. The speed analyzer 172 analyzes the speed of the moving object having an irregular movement direction to determine whether the speed is variable or not, or is greater than or equal to a reference value. The bug determination unit 173 determines that the insect is a bug when the direction of movement is irregular and the speed is variable or is higher than the reference value.

FIG. 8 is a detailed block diagram of the worm detection means 200 according to the sixth embodiment of FIG. 1, after detecting a moving object and detecting a worm by comparing with a worm shape. The bug detecting means 200 of FIG. 8 includes an image comparator 181, a candidate group generator 182, a motion detector 183, a database 184, a learner 185, a judgment criterion generator 186, and a shape. Comparing unit 187 and bug detection unit 188 is included.

The image comparator 181 compares the image of the previous frame with the image of the current frame. The candidate group generation unit 182 generates a candidate group of moving objects with respect to the part having a difference as a result of the image comparison. The motion detector 183 detects a moving object having a predetermined size or less from the generated candidate group. The database 184 stores images that are not worms and images that are worms. The learner 185 learns the presence or absence of bugs by characterizing information such as brightness difference and edge of a specific area in the database 184. The determination criteria generation unit 186 generates a determination criterion that can distinguish the presence or absence of bugs as a learning result. The shape comparison unit 187 compares the shape of the object from which the motion is detected with the shape of the determination criteria. The bug detecting unit 188 detects the object in which the motion is detected as a bug when the shape of the object in which the motion is detected and the shape of the criterion are similar.

FIG. 9 is a detailed block diagram of the worm detecting means 200 according to the seventh embodiment of FIG. 1, after detecting a moving object and detecting a worm by comparing with a worm shape. The bug detecting means 200 of FIG. 9 includes a background estimator 191, an image comparator 192, a candidate group generator 193, a motion detector 194, a database 195, a learner 196, and a criterion of determination. The generator 197, the shape comparator 198, and the insect detector 199 are included.

The background estimator 191 estimates the background of the object currently being photographed from the plurality of previous frames. The image comparator 192 compares the current frame image with the estimated background image. In other words, the background image is subtracted from the current frame image to extract an image having a difference. The candidate group generator 193 generates the image having the difference as a candidate group of the moving object based on a result of the comparison of the current frame image and the estimated background image. The motion detector 194 detects a moving object having a predetermined size or less from the generated candidate group. The database 195 stores an image which is not a worm shape and an image which is a worm shape. The learner 196 learns the presence or absence of bugs by characterizing information such as brightness difference and edge of a specific area in the database 195. The determination criteria generation unit 197 generates a determination criterion for distinguishing the presence or absence of bugs as a learning result. The shape comparison unit 198 compares the shape of the object from which the motion is detected with the shape of the determination criteria. The bug detection unit 199 detects an object in which a motion is detected as a bug when the shape of the object in which the motion is detected and the shape of the criterion are similar.

When bugs are detected, bugs may also overlap, so that a block for determining whether bugs are overlapped may be added based on information on the size, direction, and speed of the existing bugs.

The insect exterminating means 200 operates the bedbug extinguishing mechanism provided in the digital image processor according to the insect detection signal to combat the insect, and there are various methods as follows. First, there is a method of using manpower, which creates threatening motion by installing flaps around the digital image processor. This can be used to combat all kinds of insects, but this method is most effective because it is necessary to remove spider webs. Second, there is a method of using a chemical agent, which combines a chemical injector in or around the exterior of the digital image processor to spray the drug, or by applying a drug that the insect does not like on the exterior of the digital image processor to chase the insect. Third, there is a method of using smoke, which prevents insects from being approached by smoke generated by burning incense around the digital image processor. The switch may be set to generate a fire that burns incense when the insect is detected. Fourth, there is a method of using light or ultraviolet rays, which are electrically combated by radiating ultraviolet rays to the insect detection position. Fifth, there are ways to use protective gear or tattoo stickers, which are used to combat bugs by marking them on the exterior of the digital image processor. Sixth, there is a method using a sound wave, an ultrasonic wave, by installing a sound wave generator in or around the digital image processor to generate sound waves or ultrasonic waves in the wavelength band that bugs do not like to prevent bugs from approaching. Seventh, there is a method using electric shock, which generates electricity in the exterior of the digital image processor to prevent bugs from approaching. Eighth, there is a method of using a worm attractor, which installs a worm attractor in the digital image processor and destroys the worm when it is present near the digital image processor.

1 is a block diagram showing the configuration of a bug detection and eradication device in a digital image processor according to the present invention.

FIG. 2 is a detailed block diagram of a bug detecting means according to the first embodiment of FIG. 1.

3 is a detailed block diagram of a bug detecting means according to the second embodiment of FIG.

4 is a detailed block diagram of a worm detection means according to the third embodiment of FIG.

5 is a detailed block diagram of a worm detection means according to the fourth embodiment of FIG.

6 is a detailed block diagram of a bug detecting means according to a fifth embodiment of FIG. 1.

7 is a block diagram illustrating a configuration of determining a bug according to the motion detection in FIGS. 3 to 6.

8 is a detailed block diagram of a worm detection means according to the sixth embodiment of FIG.

9 is a detailed block diagram of a worm detection means according to the seventh embodiment of FIG.

FIG. 10 is a diagram illustrating an example of applying face detection and determination to insect detection and determination in FIG. 1.

Claims (11)

As a digital image processor, Worm detection means for detecting presence or absence of a worm of the target from feature extraction information of the target and / or motion information of the target; And And a worm control means for combating the detected worm by using a mechanism provided therein when the worm detection signal is received from the worm detection means. The method of claim 1, wherein the worm detection means Bug detection and eradication device characterized in that for detecting the presence / absence of insects from the voice information of the target or the moving vibration frequency of the target. The method of claim 1, wherein the worm detection means A database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A scan unit for setting a predetermined area and scanning the area from an input image; An adjusting unit for adjusting the size of the input image; And And a searcher for searching for bugs by applying the criterion of the scanned images of which the size is adjusted. The method of claim 1, wherein the worm detection means A comparison unit comparing a previous frame and a current frame; A generator configured to generate a moving object candidate group according to the comparison result; And And a detection unit for detecting the moving object having a predetermined size or less from the candidate group. The method of claim 1, wherein the worm detection means An estimator for estimating a background from previous frames; A comparison unit comparing the estimated background with a current frame; A generator configured to generate a moving object candidate group according to the comparison result; And And a detecting unit for detecting a moving object having a predetermined size or less from the candidate group. The method of claim 1, wherein the worm detection means A first comparing unit comparing the previous frame with the current frame; A first generator configured to generate a moving object candidate group according to the first comparison result; A first detector configured to detect the moving object having a predetermined size or less from the first candidate group; A second comparing unit comparing the current frame with the next frame; A second generation unit which generates a moving object candidate group according to the second comparison result; A second detector detecting the moving object having a predetermined size or less from the second candidate group; A third comparing unit comparing a position of the first detected object and a position of the second detected object; And And a third determining unit which determines that there is a movement in the closest position. The method of claim 1, wherein the worm detection means An analysis unit analyzing a shape of a movement from previous frames; A selection unit for selecting a position candidate group in which a motion is expected in a current frame from the analyzed motion type; A comparison unit comparing the candidate group with a shape of a moving object; And And a judging unit for judging that the comparison result has moved to the candidate group having the highest similarity. The method according to any one of claims 4 to 7, A first analyzer configured to analyze a movement trajectory of the object to determine whether the movement direction is irregular; A second analyzing unit which determines whether the moving direction is variable by analyzing a speed of an irregular object; And The insect detection and eradication device further comprises a determination unit for determining that the movement direction is irregular, the variable speed is an insect. The method of claim 1, wherein the worm detection means A first comparing unit comparing the previous frame with the current frame; A generator configured to generate a moving object candidate group according to the first comparison result; A first detector detecting the moving object having a predetermined size or less from the candidate group; A database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A second comparing unit comparing a shape of the detected moving object with a shape of the determination criterion; And And a second detection unit detecting a bug as a result of the second comparison. According to claim 1, wherein the worm control means An estimator for estimating a background from previous frames; A first comparing unit comparing the estimated background with a current frame; A generator configured to generate a moving object candidate group according to the first comparison result; A first detector detecting a moving object having a predetermined size or less from the candidate group; A database storing the worm image and the non-worm image; A generation unit for generating a criterion for determining a bug image through learning; A second comparison unit comparing a shape of the first detected moving object with a shape of the determination reference; And And a second detection unit detecting a bug as a result of the second comparison. According to claim 1, wherein the worm control means Insects and insects characterized in that it includes a flapper, a chemical injector, a smoke generator, a light or ultraviolet emitter, a tattoo sticker for combating insects, a sound wave or an ultrasonic generator, an electric bomber, an insect attractor, etc. Eradication device.

Images