Light information receiving method, unit and method for recognition of light-emitting objects

This application claims the benefit of Taiwan application Serial No. 98127761, filed Aug. 18, 2009, the subject matter of which is incorporated herein by reference.

1. Field of the Disclosure

The disclosure relates in general to a light information receiving method, a method and a unit for the recognition of light-emitting objects.

2. Description of the Related Art

A real-time vehicle detection and recognition system for nighttime traffic surveillance is disclosed in Taiwan Patent Application Serial No I302879. Through the light source image cutting device, light source object cutting process is performed to the captured image of the light source object. The night-time vehicle light source object recognition device uses a morpho-analysis unit to obtain the characteristics of each vehicle from the categorized light source object group. The vehicle location determination device uses a distance estimation unit to obtain the location information of the vehicles ahead and itself. After obtaining the location of the targeted light source object group, the vehicle tracking device detects the moving direction of the vehicles with respect to the continuous frames according to the location information so as to determine the information of movement of the vehicles entering the surveillance frame area.

The disclosure is directed to a light information receiving method, a method and a unit for the recognition of light-emitting objects.

According to a first aspect of the present disclosure, a light information receiving method is provided. The light information receiving method includes the following steps. A light-emitting object array is captured to obtain several images, wherein the light-emitting object array includes at least one light-emitting object. A temporal filtering process is performed to the images to recognize the light-emitting object location. A light-emitting status of the light-emitting object array is recognized according to the light-emitting object location. A decoding process is performed to output an item of information according to the light-emitting status.

According to a second aspect of the present disclosure, a light-emitting object location recognizing method is provided. The method of light-emitting object location recognizing includes the following steps. The image differencing process is performed to several images to output several differential images according to the images which are obtained by capturing a light-emitting object array having at least one light-emitting object. A logic operation is performed to output a foreground image according to the differential images. A light-emitting object location is recognized according to the foreground image.

According to a third aspect of the present disclosure, a light-emitting object location recognizing unit is provided. The light-emitting object location recognizing unit includes a storage unit, an image differencing unit, a logic unit and a location outputting unit. The image differencing unit performs image differencing according to the (i−n) th to the i th images to output several differential images. The (i−n) th to the i th images are obtained by capturing the light-emitting object array, wherein the light-emitting object array includes at least one light-emitting object. The logic unit performs logic operation according to the differential image to output the foreground image. The location outputting unit recognizes the light-emitting object location according to the foreground image. The storage unit stores a part of the images.

The disclosure will become apparent from the following detailed description of the preferred but non-limiting embodiments. The following description is made with reference to the accompanying drawings.

FIG. 1 shows a light information transmission system;

FIG. 2 shows a light emitting device;

FIG. 3 shows a flowchart of a light information receiving method;

FIG. 4 shows a foreground image under background interference;

FIG. 5 shows spatial domain information transmission;

FIG. 6 shows a light-emitting object location recognizing unit;

FIG. 7 shows a timing diagram of the temporal filter for spatial domain information transmission;

FIG. 8 shows an image differencing unit and a logic unit;

FIG. 9 shows a flowchart of step 321

FIG. 10 shows a start pattern and an end pattern;

FIG. 11 shows another light-emitting object location recognizing unit;

FIG. 12 shows another flowchart of step 321 ;

FIG. 13 to FIG. 16 respectively show three timing diagrams of the temporal filter of continuous time t˜t+3;

FIG. 17 and FIG. 18 show two timing diagrams of the temporal filter of other combinations of the start pattern and the end pattern;

FIG. 19 shows another flowchart of the light-emitting object recognizing method;

FIG. 20 shows a foreground image;

FIG. 21 shows the distance and the angle between two foreground objects;

FIG. 22 shows a first embodiment of a spatial filter;

FIG. 23 shows a second embodiment of the spatial filter;

FIG. 24 shows time domain type information transmission;

FIG. 25 shows a temporal filter for time domain information transmission; and

FIG. 26 shows another image differencing unit and another logic unit.

Light Information Transmission System:

Referring to FIG. 1 , FIG. 2 and FIG. 3 . FIG. 1 shows a light information transmission system. FIG. 2 shows a light emitting device. FIG. 3 shows a flowchart of a light information receiving method. The light information transmission system 10 includes a light emitting device 110 and a light receiving device 120 . The light emitting device 110 further includes a control circuit 112 and a light emitting device 114 . The light receiving device 120 further includes an image capturing unit 122 , a recognizing unit 124 and a decoding unit 126 . The recognizing unit 124 further includes a light-emitting object location recognizing unit 1242 and a light-emitting object status recognizing unit 1244 .

The light emitting device 114 is a light-emitting object array having at least one light-emitting object for example. That is, the light-emitting object array may include one single light-emitting object or several light-emitting objects. The light-emitting object being in bright status denotes “1” signal. To the contrary, the light-emitting object being in dark status denotes “0” signal. Besides, the transmitted signal translated by the light emitting device 114 can be the patterns with different combinations of arrangements. The long and the short signals of the time axis denote texts with meaning. The light emitting device 114 of FIG. 1 is exemplified as a 4×2 light-emitting object 1142 in FIG. 2 , and the pattern of the light-emitting object 1142 has 2 8 combinations. The control circuit 112 controls the light emitting device 114 according to an information D, so that the information D can be transmitted with light through the light emitting device 114 .

The light information receiving method, adapted to the light receiving device 120 , includes an image capturing step 310 , a light-emitting object location recognizing step 320 , a light-emitting object status recognizing step 330 and a decoding step 340 . Firstly, in the image capturing step 310 , the image capturing unit 122 captures the image of the light-emitting object 1142 of the light emitting device 114 , wherein the image capturing unit 122 is a video recorder or a camera for example.

Next, the method proceeds to the light-emitting object location recognizing step 320 , the light-emitting object location recognizing unit 1242 recognizes the locations of all light-emitting objects in the image. In the light-emitting object location recognizing step, the light-emitting object location is recognized with the information of known transmission protocol, such as the start and the end pattern of each item of information, the geometric relation of the light-emitting object and the transmission frequency. Then, the method proceeds to the light-emitting object status recognizing step 330 , the light-emitting object status recognizing unit 1244 recognizes the light-emitting status of the light-emitting object 1142 according to the light-emitting object location.

On the part of image recognition, under a complicated background environment, it is very difficult to extract the light-emitting object and recognize its status of brightness or darkness from each image by image processing technology. This is because the employed technology is not other than the foreground and background extraction technique. However, a background image suitable for image processing cannot be found, so some advanced image processing technologies such as image morphology and topology are employed. Furthermore, color image processing technology may be employed. Under such circumstance, the computer processing speed will be slowed down.

Therefore, the light receiving device 120 recognizes the light-emitting object location by the light-emitting object location recognizing unit 1242 . Then, partial image with light-emitting object pattern is recognized for the recognition of the light-emitting object status with respect to the light-emitting object location through the light-emitting object status recognizing unit 1244 . The light-emitting object status recognizing unit 1244 only needs to recognize partial image having light-emitting object pattern instead of recognizing the whole image, so the status of the light-emitting object can be easily and promptly recognized without employing complicated image processing technology.

After the light-emitting object status is recognized, the method proceeds to the decoding step 340 , the decoding unit 126 performs a decoding process to output an item of information D according to the light-emitting status of the light emitting device 114 . Thus, the information D of the light emitting device 110 is transmitted to the light receiving device 120 in the form of light signal so as to achieve the transmission of information.

Referring to FIG. 3 and FIG. 4 . FIG. 4 shows a foreground image under surrounding interference. The light-emitting object location recognizing step 320 disclosed above is itself a light-emitting object location recognizing method. The light-emitting object location recognizing step 320 further includes sub-steps 321 to 323 . Firstly, as indicated in step 321 , a temporal filtering process is performed to several images obtained by capturing the light-emitting object array to recognize the location of the light-emitting object 1142 in the image. The process of information transmission is easily interfered with by background environment, so the light-emitting object cannot be recognized properly. For example, in FIG. 4 , the foreground image includes the pattern 1510 of the real light-emitting object as well as other foreground objects 1520 such as other moving objects and light emitting sources. To avoid being interfered with by the background environment, the light-emitting object location recognizing unit 1242 performs a temporal filtering process to recognize the real light-emitting object location with a temporal filter.

Next, as indicated in step 322 , whether the light-emitting object location in the current image has changed is determined. If the light-emitting object location in the current image does not change, then the method proceeds to the light-emitting object status recognizing step 330 . The light-emitting object status recognizing unit 1244 recognizes the light-emitting status of the light-emitting object 1142 of the current image according to the light-emitting object location. As the light-emitting object location in the current image does not change, the light-emitting object status recognizing unit 1244 is able to recognize the light-emitting object status in the current image according to the previous location of the light-emitting object. To the contrary, if the light-emitting object location in the current image has already changed, then the light-emitting object location needs to be updated first as indicated in step 323 , then the method proceeds to the light-emitting object status recognizing step 330 . As the light-emitting object location in the current image has already changed, the light-emitting object status recognizing unit 1244 recognizes the light-emitting object status in the current image according to the updated light-emitting object location.

Spatial Domain Information Transmission:

Referring to FIG. 2 and FIG. 5 . FIG. 5 shows spatial domain information transmission. The transmission of information can be spatial domain information transmission or time domain information transmission. According to the spatial domain information transmission, information is transmitted by the light-emitting object through the combined bright/dark patterns. In FIG. 5 , the process of information transmission includes a start pattern S and an end pattern E to distinguish each item of transmitted information, wherein T denotes the transmission time of each information, td denotes the maintaining time of each pattern, and fd=1/td represents pattern transmission frequency. Thus, the image capturing frequency fc of the image capturing unit 122 of FIG. 1 is at least greater than fd so as to avoid losing information image.

Temporal Filter for Spatial Domain Information Transmission:

Referring to FIG. 6 , FIG. 7 , FIG. 8 , FIG. 9 and FIG. 10 . FIG. 6 shows a light-emitting object location recognizing unit. FIG. 7 shows a timing diagram of the temporal filter for spatial domain information transmission. FIG. 8 shows an image differencing unit and a logic unit. FIG. 9 shows a flowchart of step 321 . FIG. 10 shows a start pattern and an end pattern. To avoid being interfered with by background environment, the light-emitting object location recognizing unit 1242 recognizes the real light-emitting object location with a temporal filter. The light-emitting object location recognizing unit 1242 at least includes a temporal filter and a storage unit 12423 , wherein the temporal filter includes an image differencing unit 12422 , a logic unit 12425 and a location outputting unit 12426 . The image differencing unit 12422 further includes subtractors 124222 and 124224 . It is noted that the number of subtractors of the image differencing unit 12422 is not limited thereto, and can be flexibly adjusted according to the number of layers in the temporal filter.

The light-emitting object 1142 of FIG. 2 sequentially generates a start pattern S, an end pattern E, a start pattern S and an end pattern E at time t−T 1 −T 2 −t d , time t−T 1 −t d , time t−T 1 and time t. There are an information transmission time T 2 between time t−T 1 −T 2 −t d and time t−T 1 −t d and an information transmission time T 1 between time t−T 1 and time t. That is, the information transmission is alternately transmitted at two different times. Therefore, in practical application, more than two information transmission times can be determined. For the convenience of disclosure, only two different information transmission times are exemplified below.

The information transmission time T 1 and the information transmission time T 2 denote two different information transmission times. The image F(t), the image F(t−T 1 ), the image F(t−T 1 −t d ) and the image F(t−T 1 −T 2 −t d ) are respectively obtained by the image capturing unit 122 of FIG. 1 by sequentially capturing the end pattern E, the start pattern S, the end pattern E and the start pattern S at time t, time t−T 1 , time t−T 1 −t d and time t−T 1 −T 2 −t d .

For the light-emitting object location to be easily recognized, the start pattern S and the end pattern E of the information are set to be complementary pattern. For the convenience of disclosure, FIG. 10 shows four examples of the complementary patterns. The start pattern S and the end pattern E can be selected from one of the complementary patterns illustrated in FIG. 10 or other complementary patterns not illustrated in FIG. 10 .

Firstly, as indicated in step 3212 , the image differencing unit 12422 performs image differencing to output a differential image K(t) and a differential image K(t−T 1 −t d ) according to the image F(t), the image F(t−T 1 ), the image F(t−T 1 −t d ) and the image F(t−T 1 −T 2 −t d ). The image F(t), the image F(t−T 1 ), the image F(t−T 1 −t d ) and the image F(t−T 1 −T 2 −t d ) are respectively obtained by capturing the light-emitting object 1142 of FIG. 2 by the image capturing unit 122 of FIG. 1 at time t, time t−T 1 , time t−T 1 −t d and time t−T 1 −T 2 −t d . The image F(t−T 1 −T 2 −t d ) and the image F(t−T 1 ) are start patterns, and the image F(t−T 1 −t d ) and the image F(t) are end patterns.

The subtractor 124222 deducts the image F(t−T 1 −T 2 −t d ) from the image F(t−T 1 −t d ) to output the differential image K(t−T 1 −t d ), and the subtractor 124224 deducts the image F(t−T 1 ) from the image F(t) to output the differential image K(t). The image F(t−T 1 ), the image F(t−T 1 −t d ) and the image F(t−T 1 −T 2 −t d ) are stored in the storage unit 12423 when the light-emitting object location recognizing unit 1242 receives the image F(t).

Next, as indicated in step 3214 , the logic unit 12425 performs a logic operation to output the foreground image FF according to the differential image K(t) and the differential image K(t−T 1 −t d ), wherein the logic operation is an intersection (AND) operation for example. Lastly, as indicated in step 3216 , the location outputting unit 12426 recognizes the light-emitting object location of the light-emitting object 1142 of FIG. 2 in the image according to the foreground image FF.

Referring to FIG. 7 , FIG. 11 and FIG. 12 . FIG. 11 shows another light-emitting object location recognizing unit. FIG. 12 shows another flowchart of step 321 . The light-emitting object location recognizing unit 1242 further includes a gray unit 12421 , a binarization unit 12424 and a denoising unit 12427 . Step 321 further includes sub-steps 3211 , 3213 and 3215 . If the image outputted by the image capturing unit 122 of FIG. 1 is a gray image, then step 3211 can be omitted. To the contrary, if the image outputted by the image capturing unit 122 of FIG. 1 is a color image, then, as indicated in step 3211 , the gray unit 12421 grays the image F(t), the image F(t−T 1 −T 2 −t d ), the image F(t−T 1 −t d ) and the image F(t−T 1 ) to output corresponding gray images. Then, the image differencing unit 12422 further performs image differencing to the gray images of the image F(t), the image F(t−T 1 −T 2 −t d ), the image F(t−T 1 −t d ) and the image F(t−T 1 ) to output the differential image K(t) and the differential image K(t−T 1 −t d ).

Afterwards, as indicated in step 3213 , the binarization unit 12424 binarizes the differential image K(t) and the differential image K(t−T 1 −t d ) to output the binarized images. Then, the logic unit 12425 performs a logic operation to the binarized images of the differential image K(t) and the differential image K(t−T 1 −t d ) to output a logic operation result. After that, as indicated in step 3215 , the denoising unit 12427 performs a denoising process such as expansion or erosion to the logic operation result to output the foreground image FF. Then, the location outputting unit 12426 further recognizes the light-emitting object location of the light-emitting object 1142 in the image F(t) according to the foreground image FF.

Referring to FIG. 13 to FIG. 16 . FIG. 13 to FIG. 16 respectively show four timing diagrams of the temporal filter of continuous times t˜t+3. Let FIG. 15 be taken for example, the number of foreground objects in the foreground image FF equals 4, and the number of light-emitting objects equals 8. If the number of foreground objects is smaller than that of light-emitting objects, this implies that the image F(t) is not the end pattern of the transmission information. Meanwhile, new images must be captured for the recognition of the light-emitting object location.

To the contrary, if the number of foreground objects is greater than that of light-emitting objects, then other images of previous layer is obtained for performing temporal filtering. For example, if the number of foreground objects of FIG. 16 equals 10, and the number of foreground objects of FIG. 14 equals 8, then other images can be obtained from the previous layer for performing temporal filtering process to recognize the real light-emitting object location. As only two layers within temporal filter are set in the embodiments of FIG. 14 and FIG. 16 , the current recognition process of the temporal filter terminates here. Therefore, the present embodiment can recognize the location of the real light-emitting object by increasing the layer of the temporal filter.

Referring to FIG. 17 and FIG. 18 . FIG. 17 and FIG. 18 show two timing diagrams of the temporal filter of other combinations of the start pattern and the end pattern. Except for the foregoing illustration in FIG. 16 , the temporal filter can perform temporal filtering with the combination of the start pattern and the end pattern to recognize the real light-emitting object location as indicated in FIG. 17 and FIG. 18 .

Spatial Filter for Spatial Domain Information Transmission:

Referring to FIG. 19 and FIG. 20 . FIG. 19 shows another flowchart of the light-emitting object recognizing method. FIG. 20 shows a foreground image. In addition to performing temporal filtering with a temporal filter to recognize the real light-emitting object location, the light-emitting object location recognizing unit 1242 of FIG. 1 can further perform spatial filtering with a spatial filter to recognize the real light-emitting object location. That is, the light-emitting object recognizing method of FIG. 19 differs with the light-emitting object recognizing method of FIG. 3 in that the light-emitting object recognizing method of FIG. 19 further includes step 324 . According to step 324 , the light-emitting object location recognizing unit 1242 performs a spatial filtering process according to the result of temporal filtering with a spatial filter to recognize the real light-emitting object location. The spatial filter recognizes the light-emitting object location according to the geometric arrangement relation of the light-emitting object. The geometric arrangement relation is the shape, the arrangement pattern, the central point location, the mutual distance or the slope relation of the light-emitting object for example. Let FIG. 20 be taken for example, the lengths d 1 ˜d 7 denote the shortest distance between each foreground object and its adjacent foreground object. For example, the foreground object 1 is closest to the foreground object 4 , and the distance between the foreground object 1 and the foreground object 4 is the length d 1 . The angles s 1 ˜s 7 respectively denote the angles of the lengths d 1 ˜d 7 .

Referring to FIG. 21 , the distance and the angle between two foreground objects are shown. The coordinates of the two foreground objects are respectively (x1, y1) and (x2, y2). The lengths d 1 ˜d 7 are obtained from the equation d=√{square root over ((x1−x2) 2 +(y1−y2) 2 )}{square root over ((x1−x2) 2 +(y1−y2) 2 )}, and the angles s 1 ˜s 7 are obtained from the equation

Referring to FIG. 22 , a first embodiment of a spatial filter is shown. In FIG. 22 , the lengths d 1 ˜d 8 denote the shortest distance between each foreground object and its adjacent foreground objects. The angles s 1 ˜s 8 respectively denote the angles between the lengths d 1 ˜d 8 and the horizontal axis. For example, the length d 1 denotes the distance between the foreground object 1 and the closest foreground object 2 . As the lengths d 1 ˜d 6 are almost the same and the angles s 1 ˜s 6 are almost equal to 90 degrees, based on the statistical analysis of the lengths d 1 ˜d 8 and the angles s 1 ˜s 8 , it is concluded that the foreground objects 1 ˜ 8 belong to the same group, and the foreground object 9 and the foreground object 10 are other interfering objects. Also, whether the relation of the foreground objects 1 ˜ 8 conforms to the geometric arrangement relation of the real light-emitting object is determined. In the present embodiment, the relation of the foreground objects 1 ˜ 8 conforms to the geometric arrangement relation of the real light-emitting object.

Referring to FIG. 23 , a second embodiment of the spatial filter is shown. In FIG. 22 , the lengths d 1 ˜d 7 denote the shortest distance between each foreground object and its adjacent foreground object. The angles s 1 ˜s 7 respectively denote the angles between the lengths d 1 ˜d 7 and the horizontal axis. For example, the length d 1 denotes the distance between the foreground object 1 and the closest foreground object 4 . As the lengths d 1 ˜d 6 are different and the angles s 1 ˜s 6 are not exactly the same, based on the statistical analysis of the lengths d 1 ˜d 7 and the angles s 1 ˜s 7 , it is concluded that the foreground objects 1 ˜ 8 do not belong to the same group, and the relation of the foreground objects 1 ˜ 8 does not conform to the geometric arrangement relation of the real light-emitting object.

Time Domain Information Transmission:

Referring to FIG. 24 , time domain information transmission is shown. For the convenience of elaborating the light-emitting object recognizing method for time domain information transmission, the light-emitting object of the following embodiment is exemplified by a light emitting source. In FIG. 24 , the light-emitting object transmits information through the time difference relation of the bright status and the dark status. The process of information transmission includes a start period TS and an end period TE to distinguish each item of transmitted information. The light-emitting object sequentially generates three bright-dark-bright patterns within the start period TS and sequentially generates the bright-dark-bright patterns within the end period TE. There are three time intervals D 1 , D 2 and D 3 between the start period TS and the end period TE. The time intervals D 1 , D 2 and D 3 denote the transmitted information, wherein T denotes the transmission time of each information, t d denotes the maintaining time of each pattern, and fd=1/t d equals pattern transmission frequency. Thus, the image capturing frequency fc of the image capturing unit 122 of FIG. 1 is at least greater than fd so as to capture information image.

When the light information transmission system operates under an ideal background environment, the location of the light-emitting object in the image can be obtained by simple foreground image capturing technology. Meanwhile, the reference image when the light-emitting object is dark can also be recognized. Then, the status of the light-emitting object is recognized with respect to each frame, and the time interval between this time when the light-emitting object turns to bright status from dark status and the next time when the light-emitting object turns to bright status from dark status again is calculated.

The start period TS, the time intervals D 1 , D 2 and D 3 and the end period TE can be measured through the image capturing time of the image capturing unit 122 of FIG. 1 . Lastly, the transmitted information can be obtained according to the time intervals D 1 , D 2 and D 3 and the decoding step. Likewise, if there are more than two light-emitting objects, then more than two groups of time intervals D 1 , D 2 and D 3 will be employed to denote the transmitted information.

Temporal Filter for Time Domain Information Transmission:

Referring to FIG. 25 and FIG. 26 . FIG. 25 shows a temporal filter for time domain information transmission. FIG. 26 shows another image differencing unit and another logic unit. Normally, information transmission is interfered with by background environment. According to the principles of temporal filter, the start period TS and the end period TE are known. Likewise, the recognition of the light-emitting object location can be performed according to image processing such as image differencing, binarization, denoising operation and intersection operation.

The information transmission time T 1 and the information transmission time T 2 denote two different information transmission times. That is, the information transmission is alternately transmitted at two different times. More than two different times of information transmission can be set in practical application. For the convenience of disclosure, only two different information transmission times are exemplified below. The time t d denotes the maintaining time of each pattern. In the present embodiment, TC_ 1 equals T 1 −TS−TE, and TC_ 2 equals T 2 −TS−TE.

The light-emitting object sequentially generates bright-dark-bright patterns within a start period TS, and the image capturing unit 122 of FIG. 1 respectively captures three bright status-dark-bright patterns at time t−TE−TC_ 1 , time

t - TE - TC_ ⁢ 1 - TS 2

and time t−TE−TC_ 1 −TS to output the image F(t−TE−TC_ 1 ), the image

F ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 )

and the image F(t−TE−TC_ 1 −TS). The light-emitting object sequentially generates three in bright-dark-bright patterns within the end period TE, and the image capturing unit 122 of FIG. 1 respectively captures three bright-dark-bright patterns at time t−TE, time

t - TE 2

and time t to output the image F(t−TE), the image

F ⁡ ( t - TE 2 )

and the image F(t).

The image differencing unit 12422 further includes subtractors 124222 , 124224 , 124226 and 124228 . The number of subtractors of the image differencing unit 12422 is not limited to four, and can be flexibly adjusted according to the number of layers in the temporal filter. The subtractor 124222 deducts the image F(t−TE−TC_ 1 −TS) from the image

F ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 )

to output the differential image

K ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 ) .

The subtractor 124224 deducts the image

F ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 )

from the image F(t−TE−TC_ 1 ) to output the differential image K(t−TE−TC_ 1 ). The subtractor 124226 deducts the image F(t−TE) from the image

F ⁡ ( t - TE 2 )

to output the differential image

K ⁡ ( t - TE 2 ) .

The subtractor 124228 deducts the image

F ⁡ ( t - TE 2 )

from the image F(t) to output the differential image K(t). The logic unit 12425 performs intersection operation to the differential image

K ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 ) ,

the differential image K(t−TE−TC_ 1 ), the differential image

K ⁡ ( t - TE 2 )

and the differential image K(t) to output the foreground image FF.

In other words, as the image F(t) and the image

F ⁡ ( t - TE 2 )

respectively are the images of the light-emitting object in bright status and dark status, the image F(t) and the image

F ⁡ ( t - TE 2 )

for capturing the foreground image such as the differential image K(t) are obtained through image differencing. Likewise, the differential image

K ⁡ ( t - TE 2 ) ,

the differential image K(t−TE−TC_ 1 ) and the differential image

K ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 )

can be obtained according to the image

F ⁡ ( t - TE 2 ) ,

the image F(t−TE), the image F(t−TE−TC_ 1 ), the image

F ⁡ ( t - TE - TC_ ⁢ 1 - TS 2 )

and the image F(t−TE−TC_ 1 −TS). Then, the location of the foreground object can be obtained through intersection operation.

If the number of foreground objects is smaller than that of light-emitting objects, this implies that the image F(t) is not the end image of transmission information, and new image must be captured for the recognition of the light-emitting object location. If the number of foreground objects is greater than or equal to the number of light-emitting objects, then the number of layers in the temporal filter can be further increased for the recognition of the light-emitting object location. Besides, extra spatial filter can be added to assist the recognition of the light-emitting object location.

The light information receiving method, the light-emitting object location recognizing method and the light-emitting object recognizing unit disclosed in the above embodiments of the disclosure have many advantages exemplified below:

Firstly, the interference from environmental background is avoided

Secondly, no complicated image processing is required.

Thirdly, the light-emitting object status can be correctly recognized.

While the disclosure has been described by way of example and in terms of a preferred embodiment, it is to be understood that the disclosure is not limited thereto. On the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.