TWI577189B - Methods and apparatus for adjusting auxiliary lamp during image capturing - Google Patents

Description

Method and equipment for adjusting auxiliary light source in image capturing Set

The present invention relates to the field of image capture technology, and in particular, to a method and apparatus for adjusting an auxiliary light source in image capture.

"Machine vision" is the use of machines instead of human eyes for measurement and judgment. Machine vision system refers to capturing images through machine vision products (ie, image capture devices, divided into cmos and ccd), and then transmitting the images to a processing unit for digital processing, size, shape, and color. The discrimination of the device, and further, the information of the field device is obtained based on the result of the discrimination.

Machine vision is more and more widely used in automated production, including barcode recognition, character recognition, target object positioning, color detection, etc. Different application scenarios have different requirements for images.

Among the same ambient light, the adjustment of the optical parameters of the lens can not meet all application requirements. Therefore, it is necessary to set up an auxiliary light source to meet the different application requirements by setting parameters of multiple sets of auxiliary light sources.

At present, in the setting of the auxiliary light source, the manual adjustment mode is mostly used.

The manual adjustment method has the following disadvantages: Manual adjustment of lighting operation is cumbersome and unsatisfactory. Once the ambient light source changes, the operator needs to recalibrate the light, and there is no judgment standard. There is no record of quantitative data, which is not conducive to the inheritance of experience.

The invention provides a method for adjusting an auxiliary light source in image capturing, which can realize automatic adjustment of the auxiliary light source, save manpower and improve efficiency.

The invention provides a device for adjusting an auxiliary light source in image capturing, which can realize automatic adjustment of the auxiliary light source, save manpower and improve efficiency.

The method for adjusting the auxiliary light source in the image capturing method comprises: setting image feature parameters; setting the light source parameter, adjusting the auxiliary light source according to the light source parameter; performing image capturing to obtain the captured image; determining the image of the image; The feature value corresponding to the feature parameter determines whether the feature value satisfies the preset condition. If yes, the currently set light source parameter is taken as the optimal light source parameter; otherwise, the step of performing the setting of the light source parameter is returned.

Preferably, a correspondence between a target object type and the image feature parameter is established in advance, and the setting the image feature parameter includes: determining a current target object type, and selecting the target object type corresponding to the corresponding relationship The image feature parameter is used as the currently set image feature parameter.

Preferably, in the correspondence, when the target object type is a barcode recognition or a character recognition, the corresponding image feature parameter includes a definition and a contrast; and when the target object type is a target object positioning The corresponding image feature parameter includes the resolution; when the target object type is a color detection, the corresponding image feature parameter includes a color saturation.

Preferably, if the set image feature parameter is two or more, respectively determining an optimal light source parameter corresponding to each image feature parameter, and selecting one of the determined optimal light source parameters as the final result. The parameter, or the average value of each of the optimal light source parameters is taken as the best light source parameter of the final result.

Preferably, the method further comprises: setting a median search interval, represented as (low, high), the median search interval is a selection range of the light source parameter; and f(x) is used when the light source parameter is x; Taking the feature value of the image, and determining whether the feature value satisfies the preset condition includes: if f((low+high)/2+a)<=f((low+high)/2) and f((low+high) )/2-a)<=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number; if f((low+high)/2+ a)>f((low+high)/2), then modify the median search interval to ((low+high)/2, high), return to the step of performing the setting of the light source parameter; if f((low) +high)/2+a)<f((low+high)/2), then modify the median search interval to (low, (low+high)/2), return to perform the setting of the light source parameter step.

The device for adjusting the auxiliary light source in the image capturing, the device comprises a characteristic parameter setting unit, a light source adjusting unit, an imaging unit, a storage unit and an analyzing unit; the characteristic parameter setting unit sets the image characteristic parameter The light source adjusting unit sets a light source parameter, adjusts the auxiliary light source according to the light source parameter, and sends an imaging instruction to the imaging unit; the imaging unit receives an imaging instruction, performs image capturing, and obtains the captured image. Stored in the storage unit; the storage unit is configured to store an image captured by the imaging unit; the analyzing unit reads an image in the storage unit and an image feature parameter in the feature parameter setting unit; Determining the feature value corresponding to the image feature parameter of the image, determining whether the feature value satisfies the preset condition, and if yes, setting the currently set light source parameter as the optimal light source parameter; otherwise, sending a start command to the light source adjusting unit, again Make the settings of the light source parameters.

Preferably, the feature parameter setting unit pre-establishes and saves a correspondence relationship between a target object type and the image feature parameter; when setting the image feature parameter, determining the target object type, and selecting the corresponding relationship The image feature parameter corresponding to the target object type is used as the currently set image feature parameter.

Preferably, in the correspondence, when the target object type is a barcode recognition or a character recognition, the corresponding image feature parameter includes a definition and a contrast; when the target object type is a target object positioning Corresponding image feature parameters include the resolution; when the target object type is a color detection, the corresponding image feature parameter includes a color saturation.

Preferably, the feature parameter setting unit sets the image feature parameter to two or more; the analyzing unit respectively determines the optimal light source parameter corresponding to each image feature parameter, and selects one of the determined optimal light The source parameter is used as the final result of the optimal source parameter, or the average of each of the optimal source parameters is used as the final result of the optimal source parameter.

Preferably, the light source adjusting unit further sets a median search interval, represented as (low, high), the median search interval is a selection range of the light source parameter; and the analysis unit uses f(x) to represent the light source parameter. When the feature value of the image captured by x is used to determine whether the feature value satisfies the preset condition, specifically: if f((low+high)/2+a)<=f((low+high)/2) and f((low+high)/2-a)<=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number; if f(( Low+high)/2+a)>f((low+high)/2), then modify the median search interval to ((low+high)/2, high), send a start to the light source adjustment unit Command, set the light source parameter again; if f((low+high)/2+a)<f((low+high)/2), modify the median search interval to (low,(low+ High)/2), the start command is sent to the light source adjusting unit, and the setting of the light source parameter is performed again.

It can be seen from the above technical solution that, in the present invention, the image feature parameters are set; the light source parameters are set, the auxiliary light source is adjusted according to the light source parameters; the image capturing is performed to obtain the captured image; and the image characteristic parameters corresponding to the image are determined. The feature value determines whether the feature value satisfies the preset condition. If yes, the currently set light source parameter is taken as the optimal light source parameter. Otherwise, the step of setting the light source parameter is returned until the optimal light source parameter is determined. According to the technical solution of the present invention, the light source parameters of the auxiliary light source are adaptively adjusted based on the feature values of the captured image until the optimal light source parameter is determined; thereby, the automatic auxiliary light source adjustment is realized, and manpower is saved. Improve work efficiency.

10‧‧‧Characteristic parameter setting unit

20‧‧‧Light source adjustment unit

30‧‧‧ imaging unit

40‧‧‧Processing unit storage unit

50‧‧‧Analysis unit

101~104, 201~205, 301~303‧‧‧ steps

1 is a schematic flow chart of a method for adjusting an auxiliary light source in image capture according to the present invention; FIG. 2 is a flow chart of a method for adjusting an auxiliary light source in image capture according to the present invention; FIG. 3 is a light source parameter and image feature value according to the present invention; FIG. 4 is a first example of an image captured by an auxiliary light source according to the present invention; FIG. 5 is an example 2 of an image captured by an auxiliary light source according to the present invention; FIG. Example 3 of the image; and FIG. 7 is a schematic structural diagram of an apparatus for adjusting the auxiliary light source in the image capturing of the present invention.

In order to make the objects, technical solutions and advantages of the present invention more comprehensible, the present invention will be further described in detail below in conjunction with the embodiments and drawings.

The existing scheme uses manual mode to adjust the auxiliary light source. The operation is cumbersome, the accuracy is not good, and there is no judgment standard. There is no record of quantitative data, which is not conducive to the experience inheritance. In view of this, the present invention provides a scheme for adaptively adjusting the auxiliary light source in combination with the feature values of the captured images, thereby saving manpower and improving work efficiency. Referring to FIG. 1 , an auxiliary light source is used in image capturing according to the present invention. A schematic flow chart of a method of adjustment comprising the steps of:

Step 101: Set image feature parameters.

For an image, the image feature parameters can be used to indicate its characteristics. The image feature parameters are, for example, sharpness, contrast, color saturation, etc. The feature value corresponding to the image feature parameter can be used to know whether the image is the best effect that needs to be taken.

Step 102: setting a light source parameter, and adjusting the auxiliary light source according to the light source parameter.

In step 103, image capturing is performed to obtain the captured image.

The target object is imaged in the external ambient light provided by the auxiliary light source.

Step 104: Determine the feature value corresponding to the image feature parameter of the image, determine whether the feature value satisfies the preset condition, and if yes, set the currently set light source parameter as the optimal light source parameter; otherwise, return to step 102.

After capturing the image, the feature value corresponding to the image feature parameter set in step 101 can be obtained. For example, the image feature parameter is color saturation, and the corresponding feature value is 9263.

In this step, after determining the feature value corresponding to the image feature parameter of the image, it is determined whether the feature value satisfies the preset condition, and the preset condition can be set according to requirements, and the setting manner is various. For example, setting with the empirical value, correspondingly, determining whether the feature value satisfies the empirical value requirement (may be specifically a range of feature values determined by experience), and if so, determining the currently set light source parameter as the optimal light source parameter, otherwise After adjusting the light source parameters, continue shooting. For another example, using a recursive median search method to perform a discriminating query to determine The optimal light source parameter, specifically, the median search interval is set to be (low, high), the median search interval is the selection range of the light source parameter, and f(x) is used to indicate the feature of the captured image when the light source parameter is x. a value, wherein determining whether the feature value satisfies a preset condition comprises: if f((1ow+high)/2+a)<=f((low+high)/2) and f((low+high)/2- a) <=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number; if f((low+high)/2+a)>f ((low+high)/2), the median search interval is modified to ((low+high)/2, high), returning to step 102; if f((1ow+high)/2+a)<f ((low+high)/2), the median search interval is modified to (low, (low+high)/2), and the process returns to step 102.

In the invention, setting image feature parameters; setting light source parameters, adjusting the auxiliary light source according to the light source parameters; performing image capturing to obtain the captured image; determining the feature value corresponding to the image feature parameter of the image, and determining whether the feature value is satisfied The preset condition, if yes, takes the currently set light source parameter as the optimal light source parameter, otherwise, returns to step 102 of performing the set light source parameter until the optimal light source parameter is determined. According to the solution of the invention, the light source parameters of the auxiliary light source are adaptively adjusted based on the feature values of the captured image until the optimal light source parameter is determined; thereby, the automatic auxiliary light source adjustment is realized, saving manpower and improving Work efficiency.

In the present invention, different image feature parameters can be set for different target object types, and correspondingly, the correspondence between the target object type and the image feature parameters is established in advance. In step 1 of FIG. 1 , the setting of the image feature parameters specifically includes: The current target object type is determined, and the image feature parameter corresponding to the target object type is selected in the corresponding relationship as the currently set image feature parameter.

The following is an example of the specific content of the corresponding relationship. See the schematic diagram given in Table 1. When the target object type is barcode recognition or character recognition, the corresponding image feature parameters include sharpness and contrast; when the target object type is target object positioning Corresponding image feature parameters include sharpness; when the target object type is color detection, the corresponding image feature parameters include color saturation.

If the image feature parameters set in step 101 are two or more, the optimal light source parameters corresponding to each image feature parameter may be separately determined, and selected. Determine one of the best source parameters as the best source parameter for the final result, or use the average of each optimal source parameter as the best source parameter for the final result.

The method for adjusting the auxiliary light source in the image capturing of the present invention is described below by using the flow of FIG. 2, which includes the following steps:

Step 201: Establish a correspondence between the target object type and the image feature parameter.

Step 202: Determine a current target object type, and select an image feature parameter corresponding to the target object type in the corresponding relationship as the currently set image feature parameter.

In step 203, a median search interval is set, a light source parameter is selected in the median search interval, and the auxiliary light source is adjusted according to the light source parameter.

The median search interval is represented as (low, high), the median search interval is the selection range of the light source parameters, low is the lowest value, and high is the maximum value. Corresponding to the specific example given in Figure 3, initially, the value of low is 0 and the value of high is MAX.

In this example, the light source parameters are respectively set to x=(low+high)/2, x=(low+high)/2+a, and x=(low+high)/2-a, respectively, for image capturing. a takes a value of 1.

Step 204: Perform image capturing to obtain the captured image.

Step 205: Determine the feature value corresponding to the image feature parameter of the image, and determine whether the feature value satisfies the preset condition. If yes, use the currently set light source parameter as the optimal light source parameter; otherwise, return to step 203.

Use f(x) to represent the eigenvalue of the image captured when the source parameter is x. In this example, the light source parameters are respectively set to x=(low+high)/2, x=(low+high)/2+a, x=(low+high)/2-a, and the corresponding eigenvalues are obtained. , denoted as f(low+high)/2, f((low+high)/2+a), and f((low+high)/2-a), respectively.

The foregoing step is to determine whether the feature value satisfies a preset condition, and specifically includes: if f((low+high)/2+a)<=f((low+high)/2) and f((low+high)/ 2-a)<=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number, in this example a is 1; if f(( Low+high)/2+a)>f((low+high)/2), then modify the median search interval to ((low+high)/2, high), return to step 203; if f(( If low+high)/2+a)<f((low+high)/2), the median search interval is changed to (low, (low+high)/2), and the process returns to step 203.

Here, the target object type is specifically detected by color, and an example is illustrated in conjunction with FIG. 4-6. First, the set image feature parameters are queried from Table 1 as color saturation. The range of light source parameters: 0~255, adaptively select the light source parameters, adjust the auxiliary light source to find the best color saturation, and then determine the corresponding optimal light source parameters; the following is the process of adaptive adjustment:

1. Turn on the adjustable auxiliary light source and imaging device.

2. Run the visual analysis program, the program will set the light source parameter to 128 first, take a picture and take image (as shown in Figure 4), and after analysis, the color saturation of the photo is 8387.

3. The visual analysis program sets the light source parameters to 127 and 129 respectively, and takes a picture (as shown in Figure 5). After analysis, the color saturation of the photos is 8372 respectively. And 8936, the saturation value shows an upward trend, indicating that the best saturation has not been found.

4. Visual analysis program According to the calculation, by setting different light source parameters, taking pictures, analyzing the saturation value, and finally setting the light source parameters to 195, 196 and 197 respectively, taking pictures (as shown in Figure 6), and obtaining photos. The color saturations were 9263, 9276 and 9272, respectively, and the best source parameter was found to be 196.

Referring to FIG. 7, a schematic structural diagram of an apparatus for adjusting an auxiliary light source in image capturing according to the present invention is shown. The auxiliary light source adjusting device includes a feature parameter setting unit 10, a light source adjusting unit 20, an imaging unit 30, a storage unit 40, and an analyzing unit 50. The image parameter setting unit 10 sets the image feature parameter; the light source adjusting unit 20 sets the light source parameter, adjusts the auxiliary light source according to the light source parameter, and sends an imaging instruction to the imaging unit 30; the imaging unit 30 receives the imaging instruction and performs image capturing. And storing the captured image in the storage unit 40; the storage unit 40 is configured to store the image captured by the imaging unit 30; the analyzing unit 50 reads the image in the storage unit 40 and the image in the feature parameter setting unit 10 a characteristic parameter; determining a feature value corresponding to the image feature parameter of the image, determining whether the feature value satisfies a preset condition, and if yes, using the currently set light source parameter as an optimal light source parameter; otherwise, sending a start command to the light source adjusting unit 20 , set the light source parameters again.

Preferably, the feature parameter setting unit 10 pre-establishes and saves the correspondence relationship between the target object type and the image feature parameter; when setting the image feature parameter, determines the target object type, and selects an image corresponding to the target object type in the corresponding relationship. The feature parameter is used as the currently set image feature parameter.

Preferably, in the correspondence relationship: when the target object type is barcode recognition or character recognition, the corresponding image feature parameters include sharpness and contrast; when the target object type is target object positioning, the corresponding image feature parameters include definition When the target object type is color detection, the corresponding image feature parameters include color saturation.

Preferably, the feature parameter setting unit 10 sets the image feature parameters to two or more; the analyzing unit 50 respectively determines the optimal light source parameters corresponding to the image feature parameters, and selects one of the determined optimal light source parameters as the final result. The best source parameters, or the average of the best source parameters as the best source parameter for the final result.

Preferably, the light source adjusting unit 20 further sets a median search interval, which is represented as (low, high), and the median search interval is a selection range of the light source parameter; and the analyzing unit 50 uses f(x) to indicate that the light source parameter is x. When the feature value of the captured image is used to determine whether the feature value satisfies the preset condition, specifically: if f((low+high)/2+a)<=f((low+high)/2) and f(( Low+high)/2-a)<=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number; if f((low+high) /2+a)>f((low+high)/2), the median search interval is modified to ((low+high)/2, high), a start command is sent to the light source adjustment unit 20, and the light source parameter is again performed. Setting; if f((low+high)/2+a)<f((low+high)/2), modify the median search interval to (low, (low+high)/2), to the light source The adjustment unit 20 sends a start command to set the light source parameters again.

The invention sets image feature parameters automatically for application requirements, automatically The image is taken under different external light sources, and the optimal light source parameters are found according to the relationship between the light source parameters and the characteristic values. The inventive solution has the following advantages over manual adjustment of the light:

1, improve efficiency, compared to the cumbersome manual operation, the solution of the invention automatically completes the setting and reduces manual intervention.

2, improve accuracy, and reduce errors caused by labor.

3, easy to expand, for new application requirements, only need to set the image feature parameters for this demand, it is applicable to this program.

The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalents, improvements, etc., which are made within the spirit and principles of the present invention, should be included in the present invention. Within the scope of protection.

101~104‧‧‧Steps

Claims (6)

A method for adjusting an auxiliary light source in image capturing, comprising: setting an image feature parameter; setting a light source parameter, adjusting an auxiliary light source according to the light source parameter, performing image capturing, acquiring the captured image; determining the image a feature value corresponding to the image feature parameter, determining whether the feature value satisfies a preset condition, and if yes, using the currently set light source parameter as an optimal light source parameter; otherwise, returning to performing the setting of the light source a step of parameter; wherein, a correspondence between a target object type and the image feature parameter is pre-established, and the setting the image feature parameter includes: determining a current target object type, and selecting the target object in the correspondence relationship The image feature parameter corresponding to the type is used as the currently set image feature parameter; wherein, in the correspondence relationship, when the target object type is a barcode recognition or a character recognition, the corresponding image feature parameter includes a definition And a contrast; when the target object type is a target object, This image characteristic parameter comprises the sharpness; when the target object type is a color detection, corresponding to the image characteristic parameter includes a color saturation. The method of claim 1, wherein if the image feature parameter is set to be more than two, an optimal light source parameter corresponding to each image feature parameter is determined, and one of the determined ones is selected. The best light source parameter is used as the final result of the optimal light source parameter, or the average of each of the optimal light source parameters is used as the final result of the optimal light source parameter. The method of claim 1 or 2, wherein the method further comprises: setting a median search interval, represented as (low, high), the median search interval is a selection range of the light source parameter; f(x) represents the feature value of the captured image when the light source parameter is x, and whether the determined feature value satisfies the preset condition includes: if f((low+high)/2+a)<=f((low+ High)/2) and f((low+high)/2-a)<=f((low+high)/2), then (low+high)/2 is the best source parameter, a is a positive real number If f((low+high)/2+a)>f((low+high)/2), the median search interval is modified to ((low+high)/2, high), returning to perform The step of setting the light source parameter; if f((low+high)/2+a)<f((low+high)/2), the middle value search interval is modified to (low, (low+high)/ 2), return to the step of performing the setting of the light source parameter. A device for adjusting an auxiliary light source in image capturing, wherein the device comprises a characteristic parameter setting unit, a light source adjusting unit, an imaging unit, a storage unit and an analyzing unit; the characteristic parameter setting unit sets an image feature The light source adjusting unit sets a light source parameter, adjusts an auxiliary light source according to the light source parameter, and sends an imaging instruction to the imaging unit; The imaging unit receives the imaging instruction, performs image capturing, acquires the captured image, and stores the captured image in the storage unit; the storage unit is configured to store the image captured by the imaging unit; the analyzing unit reads the storage An image in the unit and the image feature parameter in the feature parameter setting unit; determining a feature value corresponding to the image feature parameter of the image, determining whether the feature value satisfies a preset condition, and if yes, setting the currently set The light source parameter is used as an optimal light source parameter; otherwise, a start command is sent to the light source adjusting unit, and the light source parameter is set again; wherein the feature parameter setting unit pre-establishes and saves a target object type and the image feature parameter a corresponding relationship between the image features, the target image type is determined, and the image feature parameter corresponding to the target object type is selected as the currently set image feature parameter; In the correspondence: when the target object type is a barcode recognition or a character recognition Corresponding image feature parameters include a sharpness and a contrast; when the target object type is a target object, the corresponding image feature parameter includes the sharpness; when the target object type is a color detection, The corresponding image feature parameter includes a color saturation. The device of claim 4, wherein the feature parameter setting unit sets the image feature parameter to two or more; the analyzing unit respectively determines the optimal light source parameter corresponding to each of the image feature parameters , select one of the determined optimal light source parameters as the final knot The optimal light source parameter of the fruit, or the average value of each of the optimal light source parameters is taken as the best light source parameter of the final result. The apparatus of claim 4, wherein the light source adjusting unit further sets a median search interval, which is represented as (low, high), and the median search interval is a selection range of the light source parameter; The analysis unit uses f(x) to represent the feature value of the image captured when the light source parameter is x, and determines whether the feature value satisfies the preset condition, specifically: if f((low+high)/2+a)< =f((low+high)/2) and f((low+high)/2-a)<=f((low+high)/2), then (low+high)/2 is the best The source parameter, a is a positive real number; if f((low+high)/2+a)>f((low+high)/2), the median search interval is modified to ((low+high)/2 , high), send a start command to the light source adjusting unit, and perform setting of the light source parameter again; if f((low+high)/2+a)<f((low+high)/2), The median search interval is modified to (low, (low+high)/2), the start command is sent to the light source adjustment unit, and the setting of the light source parameter is performed again.