JP7496935B2 - IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND IMAGE PROCESSING PROGRAM - Google Patents

Description

The present disclosure relates to an image processing device, an image processing method, and an image processing program.

The device receives images of various resolutions and compression methods. For example, the various resolutions are 8K, 4K, full HD (High Definition), SD (Standard Definition), etc. For example, the various compression methods are H. 265/HEVC, H. 264/AVC, MPEG2, etc. The device may be equipped with a high image quality function. It is desired that the device performs optimal image quality enhancement. For example, when sharpening a blurred image, a process for sharpening the image is executed. For example, when a noisy image is input, a process for not emphasizing the noise is executed. Here, an image quality control device that controls image quality has been proposed (see Patent Document 1). For example, the image quality control device of Patent Document 1 generates all histograms of luminance information, chromaticity information, color information, and frequency information obtained from an image included in a video signal. The image quality control device extracts histogram patterns corresponding to all histograms from a reference table having a plurality of preset histogram patterns, and controls the image quality of the video signal based on control parameters corresponding to the extracted histogram patterns.

Patent Publication No. 5651340

In the above technology, image quality is controlled using a lookup table. However, when various images are input, it is difficult to optimize image quality using only the information registered in the lookup table.

The objective of this disclosure is to optimize image quality for a variety of images.

An image processing device according to one aspect of the present disclosure is provided. The image processing device includes an acquisition unit that acquires an image and a trained model, a feature extraction unit that extracts a plurality of types of feature amounts based on the image, a correction parameter detection unit that detects a correction parameter that is a parameter for correcting the image quality of the image using the plurality of types of feature amounts and the trained model , a determination unit that determines whether or not a difference between the detected correction parameter and a preset correction parameter is equal to or less than a preset threshold, and an image processing unit that executes a process for correcting the image quality of the image using the detected correction parameter when the difference is equal to or less than the threshold .

According to the present disclosure, image quality can be optimized for a variety of images.

FIG. 2 is a diagram illustrating hardware included in the image processing device according to the first embodiment. 1 is a block diagram showing functions of an image processing device according to a first embodiment; FIG. 11 is a diagram showing an example (part 1) of horizontal and vertical edge histograms according to the first embodiment; FIG. 11 is a diagram showing an example (part 2) of horizontal and vertical edge histograms according to the first embodiment; 5A to 5C are diagrams illustrating a specific example of a method for extracting a frame difference histogram in the first embodiment. 5A to 5C are diagrams showing a specific example of a method for extracting the number of luminance changes in the first embodiment; 5A to 5C are diagrams showing a specific example of a method for extracting a saturation histogram in the first embodiment. FIG. 2 is a diagram illustrating an example of a neural network according to the first embodiment. A figure showing an example of a trained model constituting a random forest in embodiment 1. FIG. 2 is a diagram illustrating an example of a support vector machine according to the first embodiment. 4 is a flowchart showing an example of processing executed by the image processing device according to the first embodiment. FIG. 11 is a block diagram showing the functions of an image processing device according to a second embodiment. 13 is a flowchart illustrating an example of processing executed by an image processing device according to a second embodiment.

The following describes an embodiment with reference to the drawings. The following embodiment is merely an example, and various modifications are possible within the scope of this disclosure.

Embodiment 1.
1 is a diagram showing hardware included in an image processing device according to embodiment 1. The image processing device 100 is a device that executes an image processing method. The image processing device 100 includes a processor 101, a volatile storage device 102, and a non-volatile storage device 103.

The processor 101 controls the entire image processing device 100. For example, the processor 101 is a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), etc. The processor 101 may be a multiprocessor. The image processing device 100 may also have a processing circuit. The processing circuit may be a single circuit or a composite circuit.

The volatile memory device 102 is a main memory device of the image processing device 100. For example, the volatile memory device 102 is a RAM (Random Access Memory). The non-volatile memory device 103 is an auxiliary memory device of the image processing device 100. For example, the non-volatile memory device 103 is a HDD (Hard Disk Drive) or an SSD (Solid State Drive).

Next, the functions of the image processing device 100 will be described.
2 is a block diagram showing functions of the image processing device according to the first embodiment. The image processing device 100 includes a storage unit 110, an acquisition unit 120, feature extraction units 130_1, 130_2, ..., 130_n, a correction parameter detection unit 140, an image processing unit 150, and an output unit 160. Note that n is a positive integer. Here, the feature extraction units 130_1, 130_2, ..., 130_n are collectively referred to as the feature extraction unit 130.

The storage unit 110 may be realized as a storage area secured in the volatile storage device 102 or the non-volatile storage device 103 .
Some or all of the acquisition unit 120, the feature extraction unit 130, the correction parameter detection unit 140, the image processing unit 150, and the output unit 160 may be realized by a processing circuit. Also, some or all of the acquisition unit 120, the feature extraction unit 130, the correction parameter detection unit 140, the image processing unit 150, and the output unit 160 may be realized as modules of a program executed by the processor 101. For example, the program executed by the processor 101 is also referred to as an image processing program. For example, the image processing program is recorded on a recording medium.

Here, the following description will be given of the case where the image processing device 100 acquires a digital image. The image processing device 100 may acquire an analog image. When the image processing device 100 acquires an analog image, the image processing device 100 has an A (Analog)/D (Digital) converter.

The storage unit 110 may store an image and a trained model. The storage unit 110 may also store a video.
The acquisition unit 120 acquires an image. For example, the acquisition unit 120 acquires an image from the storage unit 110. Also, for example, the acquisition unit 120 acquires an image from a camera. Note that the camera is not illustrated. Also, the acquisition unit 120 acquires a video. In other words, the acquisition unit 120 acquires a plurality of images.

Furthermore, the acquisition unit 120 acquires the trained model. For example, the acquisition unit 120 acquires the trained model from the storage unit 110. Here, the trained model may be stored in an external device (e.g., a cloud server). When the trained model is stored in an external device, the acquisition unit 120 acquires the trained model from the external device.

The feature extraction unit 130 extracts multiple types of feature amounts based on the image. In other words, the feature extraction unit 130 extracts multiple different types of feature amounts based on the image. The feature extraction process will be described in detail.

The feature extraction unit 130 (e.g., feature extraction unit 130_1) extracts a horizontal edge histogram, which is one feature, based on an image. In detail, the feature extraction unit 130 calculates the luminance difference value between horizontally adjacent pixels as edge enhancement based on the luminance value of the image. For example, the feature extraction unit 130 extracts a horizontal edge histogram expressed by dividing 1024 levels of edge enhancement into 16 parts. An example of a horizontal edge histogram is shown below.

Figure 3 shows an example (part 1) of a horizontal edge histogram according to the first embodiment. The vertical axis shows the frequency of occurrence. The horizontal axis shows edge strength.

The feature extraction unit 130 extracts horizontal edge amount, which is one feature amount. Specifically, we will explain how to extract horizontal edge amount.

Figure 4 is a diagram showing an example (part 2) of a horizontal edge histogram in embodiment 1. It is assumed that the horizontal edge histogram shown in Figure 4 has been extracted. The feature extraction unit 130 extracts the total occurrence frequency between the minimum value (edge_left in Figure 4) and the maximum value (edge_right in Figure 4) of the edge strength in a specified range in the horizontal edge histogram as the horizontal edge amount. Note that the range may be a preset range.

Furthermore, the feature extraction unit 130 (for example, feature extraction unit 130_2) extracts a vertical edge histogram, which is one feature, based on the image. In detail, the feature extraction unit 130 calculates the luminance difference value between adjacent pixels in the vertical direction as edge enhancement based on the luminance value of the image. For example, the feature extraction unit 130 extracts a vertical edge histogram expressed by dividing the edge enhancement of 1024 gradations into 16 parts. For example, the vertical edge histogram can be expressed as shown in FIG. 3. Therefore, FIG. 3 may be considered to be a diagram showing an example of a vertical edge histogram.

The feature extraction unit 130 extracts a vertical edge amount, which is one feature amount. The method for extracting a vertical edge amount is the same as the method for extracting a horizontal edge amount. For example, if FIG. 4 is considered as an example of a vertical edge histogram, the feature extraction unit 130 extracts the total occurrence frequency between the minimum and maximum values of edge strength in a specified range in the vertical edge histogram as the vertical edge amount.

In addition, the feature extraction unit 130 (for example, the feature extraction unit 130_3) uses the image to extract a frame difference histogram, which is one feature. A method for extracting a frame difference histogram will be specifically described.

Figure 5 shows a specific example of a method for extracting a frame difference histogram in embodiment 1. Figure 5 shows a luminance histogram based on a previously acquired image (e.g., also referred to as the n-1th frame) and a luminance histogram based on a currently acquired image (e.g., also referred to as the nth frame).

The vertical axis of these brightness histograms indicates the number of pixels. The horizontal axis of these brightness histograms indicates brightness. The brightness histograms are obtained by dividing the brightness into 16 parts based on the maximum RGB (Red, Green, Blue) values of each pixel. In other words, the brightness histograms are obtained by summarizing the brightness in 16 gradations. Note that the brightness histograms may also be summarized in 1 gradation or 8 gradations.

The feature extraction unit 130 extracts a frame difference histogram by calculating the difference between a luminance histogram based on a previously acquired image and a luminance histogram based on a currently acquired image. FIG. 5 shows a frame difference histogram. The vertical axis of the frame difference histogram indicates the number of pixels. The horizontal axis of the frame difference histogram indicates luminance. The frame difference histogram may also be called a differential luminance histogram.

In addition, the feature extraction unit 130 (for example, the feature extraction unit 130_4) uses the image to extract a frequency histogram, which is one feature. In addition, the feature extraction unit 130 may extract the number of luminance changes as a feature. A method for extracting the number of luminance changes will be specifically described.

FIG. 6 is a diagram showing a specific example of a method for extracting the number of luminance changes in the first embodiment. The vertical axis of the graph in FIG. 6 indicates the luminance value. The horizontal axis of the graph in FIG. 6 indicates pixel coordinates. FIG. 6 shows nine pixels in the horizontal direction and the luminance values of the nine pixels. The feature extraction unit 130 extracts the number of pixels whose luminance difference value between adjacent pixels in the horizontal direction is equal to or greater than a threshold value as the number of luminance changes. For example, when the difference between the luminance value of the pixel N+2 and the luminance value of the pixel N+3 is equal to or greater than a threshold value, the feature extraction unit 130 extracts the pixel N+3 as a pixel whose luminance has changed. In FIG. 6, the pixels whose luminance has changed are represented by colored circles. FIG. 6 also shows that the number of luminance changes is 5.
Furthermore, the feature extraction section 130 may extract the total number of luminance changes as the feature.

The feature extraction unit 130 (e.g., feature extraction unit 130_5) uses an image to extract a saturation histogram, which is one feature. The method for extracting a saturation histogram will be described in detail below.

Figure 7 is a diagram showing a specific example of a method for extracting a saturation histogram in embodiment 1. The vertical axis indicates the frequency of occurrence. The horizontal axis indicates saturation. For example, the feature extraction unit 130 extracts saturation based on the maximum value when the U value and V value calculated in the CIE UVW color space are replaced with absolute values. For example, the feature extraction unit 130 extracts a saturation histogram represented by dividing 512 levels of saturation into 16.

In the above, examples of features are horizontal edge amount, horizontal edge histogram, vertical edge amount, vertical edge histogram, frame difference histogram, frequency histogram, and saturation histogram. The feature may be a feature other than those mentioned above. For example, the feature may be the maximum luminance of the image, the minimum luminance of the image, the average luminance of the image, the value of the black area in the image, the value of the white area in the image, etc.

The correction parameter detection unit 140 detects correction parameters using multiple types of feature quantities and the trained model. That is, the correction parameter detection unit 140 inputs multiple types of feature quantities into the trained model, and detects correction parameters output by the trained model. The correction parameters are parameters for correcting the image quality of an image. The correction parameters may also be expressed as parameters for improving the image quality of an image.

Next, the trained model will be described. The trained model may be configured as a multi-layer neural network. An example of the neural network will be described.
Fig. 8 is a diagram showing an example of a neural network according to the first embodiment. The neural network is composed of an input layer, an intermediate layer, and an output layer. Fig. 8 shows that the number of intermediate layers is three. The number of intermediate layers is not limited to three. Furthermore, the number of neurons is not limited to the number shown in the example of Fig. 8.

A plurality of types of feature quantities are assigned to a plurality of neurons in the input layer, for example, a horizontal edge quantity is assigned to one neuron in the input layer.
For example, the correction parameter y is calculated by equation (1) based on a plurality of types of feature amounts input to a plurality of neurons.

Note that n is the number of neurons in the input layer. x1 to xn are multiple types of features. b is the bias. w is the weight. The bias b and weight w are determined by learning.

Also, s indicates a function. The function in question is called function s(a). Function s(a) is an activation function. For example, activation function s(a) may be a step function that outputs 0 if a is 0 or less, and outputs 1 if a is other than 0. Also, for example, activation function s(a) may be a ReLU function that outputs 0 if a is 0 or less, and outputs the input value a if a is other than 0, or an identity function that outputs the input value a as is, or a sigmoid function.

Note that neurons in the input layer output the input value as is. Therefore, the activation function used in neurons in the input layer can be said to be an identity function. A step function or a sigmoid function may be considered to be used in the intermediate layer. A ReLU function may be considered to be used in the output layer. Also, different functions may be used between neurons in the same layer.

Here, as described above, the weight w is determined by learning. An example of a method for calculating the weight w will be described. First, a plurality of images and correction parameters for correcting the image quality of the plurality of images are prepared. In the image quality correction process in machine learning, it is defined that noise reduction process, sharpening process, contrast improvement process, and color conversion process are performed. In addition, in the correction process, the values used in the noise reduction process, sharpening process, contrast improvement process, and color conversion process are changed to improve the image quality. In machine learning, the difference between the output value corresponding to the input pattern and the prepared correction parameters is calculated. The weight w is determined so that the difference is small.

The trained model may be configured as a random forest. An example of a trained model that configures a random forest is shown below.
9 is a diagram showing an example of a trained model constituting the random forest of the first embodiment. FIG. 9 shows a random forest 200. The random forest 200 includes decision trees 210_1, 210_2, ..., 210_n, where n is a positive integer. Here, the decision trees 210_1, 210_2, ..., 210_n are collectively referred to as the decision tree 210. The decision tree 210 is made up of a plurality of nodes. In the decision tree 210, a tree-structured classification rule is created by stacking the nodes.

Each of the multiple decision trees 210 is assigned one of multiple types of feature quantities. For example, the horizontal edge quantity is assigned to the decision tree 210_1. The feature quantity is then input to the first-layer node of the decision tree 210. For example, the horizontal edge quantity is input to the first-layer node of the decision tree 210_1. A branching condition is set for the node. For example, the branching condition is set to a condition that the horizontal edge quantity is 1000 or more. The branching condition is determined by the condition that maximizes the information gain of the input data. The information gain refers to the classification degree (also called impurity, for example) of data when classifying data into the next node of a certain node x. Maximizing the information gain means maximizing "(impurity before classification) - (impurity after classification)" at a certain node x, and means minimizing the impurity after classification. Impurity g (T) is calculated using formula (2).

g is the Gini coefficient. The Gini coefficient is an index showing whether data classification is successful or not. T indicates data input to the node. p(i|t) indicates the number of data of a certain class for the input data. c indicates the number of classes.
The information gain I G is calculated using equation (3).

Tb indicates data before classification. z indicates the number of nodes after classification. Ni indicates the number of data in node i after classification. Np indicates the number of data before classification. Di indicates the data in node i after classification.
When the difference between the impurity of the data input to node x and the impurity after classification at node x becomes large, the information gain is maximized and the data can be best classified.
In equation (3), the Gini coefficient g was used. Entropy or classification error may also be used in equation (3).

Here, as described above, a branching condition is defined for the node. The branching condition is defined by machine learning. An example of a method for calculating a branching condition in machine learning will be described. First, a plurality of images and correction parameters for correcting the image quality of the plurality of images are prepared. In the image quality correction process in machine learning, it is defined that noise reduction process, sharpening process, contrast improvement process, and color conversion process are performed. In addition, in the correction process, the values used in the noise reduction process, sharpening process, contrast improvement process, and color conversion process are changed to improve the image quality. In machine learning, the difference between the output value corresponding to the input pattern and the prepared correction parameter is calculated. The branching condition is defined so that the difference becomes small.

In the random forest 200, correction parameters are output by taking a majority vote on the results output from multiple decision trees 210.

The trained model may also be a trained model generated using a support vector machine. Thus, the trained model reflects the technology of the support vector machine. An example of such a trained model is shown below.

Figure 10 is a diagram showing an example of a support vector machine according to the first embodiment. A linear input element is used in the support vector machine. Multiple types of features are input to the support vector machine. Each of the multiple types of features is classified into two classes by the linear input element. The linear input element is calculated using equation (4).

x indicates the input data (i.e., feature). "w^T(x) + b" indicates a linear line that separates the data. Equation (5) and equation (6) are used to calculate equation (4). Equation (5) is expressed as follows:

Equation (6) is expressed as follows:

w indicates the slope of the linear input element. ε indicates a variable that weakens the constraint in equation 6. C indicates a positive regularization coefficient. t is a variable (i.e., 1 or -1) that makes "(w^T(x) + b)" a positive value. The optimal linear input element is calculated by allowing classification failure of the linear input element using ε and C.

If the input data cannot be classified into two categories using linear input elements, a kernel function may be used. By using the kernel function, the input data is expanded into a multidimensional space, and a plane that can be linearly classified using the linear input elements is calculated. The kernel function may be an RBF kernel, a polynomial kernel, a linear kernel, or a sigmoid kernel.

Here, the linear input element is calculated by machine learning. An example of a method for calculating a linear input element in machine learning will be described. First, a plurality of images and correction parameters for correcting the image quality of the plurality of images are prepared. In the image quality correction process in machine learning, it is defined that noise reduction process, sharpening process, contrast improvement process, and color conversion process are performed. In the correction process, the values used in the noise reduction process, sharpening process, contrast improvement process, and color conversion process are changed to improve the image quality. In machine learning, the difference between the output value corresponding to the input pattern and the prepared correction parameters is calculated. The linear input element is calculated so that the difference becomes small.

The image processing unit 150 uses the correction parameters to execute processing for correcting the image quality of the image.
For example, the image processing unit 150 uses the correction parameters to execute a process of reducing noise in the image. For example, the process of reducing noise is a low-pass filter process. For example, the image processing unit 150 uses the correction parameters to correct a noisy image to an image with less noise. Also, for example, the image processing unit 150 uses the correction parameters to correct an image including fine pattern noise to an image with less noise.

Also, for example, the image processing unit 150 uses the correction parameters to execute processing for sharpening the image. For example, the processing for sharpening the image is high-pass filter processing. For example, the image processing unit 150 uses the correction parameters to correct a noisy image into a sharpened image. Also, for example, the image processing unit 150 uses the correction parameters to correct a blurry image into a sharpened image.

The image processing unit 150 may use the correction parameters to perform processing such as improving the contrast of the image and converting the color of the image.

The output unit 160 outputs the corrected image. For example, the output unit 160 outputs the corrected image to a display. This allows the user to view an image of optimal image quality. Note that an illustration of a display is omitted. The output unit 160 may also output the corrected image to an external device. The output unit 160 may also output the corrected image to the memory unit 110.

Next, the process executed by the image processing device 100 will be described with reference to a flowchart.
FIG. 11 is a flowchart illustrating an example of processing executed by the image processing device according to the first embodiment.
(Step S11) The acquisition unit 120 acquires an image.
(Step S12) The feature extraction unit 130 extracts a plurality of types of feature amounts based on the image.

(Step S13) The correction parameter detection unit 140 detects correction parameters by using multiple types of feature amounts and the trained model.
(Step S14) The image processing unit 150 executes a process for correcting the image quality of the image using the correction parameters.
(Step S15) The output unit 160 outputs the corrected image.

According to embodiment 1, the image processing device 100 corrects an image using correction parameters output from the trained model. In the learning phase of the trained model, various images are input as training data, and learning is performed to output appropriate correction parameters corresponding to the various images. Through such learning, the trained model is generated. Therefore, the image processing device 100 using the trained model can optimize the image quality of various images.

In addition, multiple types of features are extracted before the multiple types of features are input to the trained model. In other words, in the trained model, a process of extracting multiple types of features from an image is not performed. Therefore, according to embodiment 1, it is possible to reduce the weight of the trained model and to speed up processing in the trained model.

Embodiment 2.
Next, a description will be given of embodiment 2. In embodiment 2, differences from embodiment 1 will be mainly described. Furthermore, in embodiment 2, description of matters common to embodiment 1 will be omitted.

Fig. 12 is a block diagram showing the functions of an image processing device according to embodiment 2. The components in Fig. 12 that are the same as those in Fig. 2 are given the same reference numerals as those in Fig. 2.
The image processing device 100 further includes a determination unit 170. A part or the whole of the determination unit 170 may be realized by a processing circuit. Also, a part or the whole of the determination unit 170 may be realized as a module of a program executed by the processor 101.
The function of the determination unit 170 will be described in detail later.

Next, the process executed by the image processing device 100 will be described with reference to a flowchart.
Fig. 13 is a flowchart showing an example of processing executed by the image processing device of the second embodiment. The processing in Fig. 13 differs from the processing in Fig. 11 in that step S13a is executed. Therefore, step S13a will be described in Fig. 13. Descriptions of the processing other than step S13a will be omitted.

(Step S13a) The judgment unit 170 judges whether the difference between the detected correction parameter and a preset correction parameter is less than or equal to a preset threshold value.

Here, the acquisition unit 120 can acquire multiple images (e.g., video) for a certain period of time. Then, the feature extraction unit 130 and the correction parameter detection unit 140 detect multiple correction parameters based on the multiple images. In this way, multiple correction parameters detected over a certain period of time are detected. For example, the preset correction parameter is an average value based on the multiple correction parameters detected over a certain period of time.

If the difference is less than or equal to the threshold, processing proceeds to step S14. If the difference is greater than the threshold, processing ends.

Here, if the image processing device 100 corrects the image when step S13a is No, a sudden change in image quality occurs. Therefore, in order to suppress the sudden change in image quality, the image processing device 100 uses the threshold value to determine whether or not to perform correction. Therefore, according to embodiment 2, the image processing device 100 can suppress the sudden change in image quality.

The features of each of the embodiments described above can be combined with each other as appropriate.

100 Image processing device, 101 Processor, 102 Volatile storage device, 103 Non-volatile storage device, 110 Storage unit, 120 Acquisition unit, 130, 130_1, 130_2, ..., 130_n Feature extraction unit, 140 Correction parameter detection unit, 150 Image processing unit, 160 Output unit, 170 Judgment unit, 200 Random forest, 210, 210_1, 210_2, ..., 210_n Decision tree.

Claims (6)

An acquisition unit that acquires an image and a trained model;
A plurality of feature extraction units extracting a plurality of types of feature amounts based on the image;
a correction parameter detection unit that detects a correction parameter, which is a parameter for correcting image quality of the image, by using the plurality of types of feature amounts and the trained model;
a determination unit that determines whether a difference between the detected correction parameter and a preset correction parameter is equal to or smaller than a preset threshold;
an image processing unit that executes a process for correcting an image quality of the image using the detected correction parameters if the difference is equal to or smaller than the threshold value ;
An image processing device comprising: The trained model is composed of a neural network, a random forest, or a support vector machine.
The image processing device according to claim 1 . the plurality of types of feature amounts are two or more types of feature amounts selected from a horizontal edge amount, a horizontal edge histogram, a vertical edge amount, a vertical edge histogram, a frame difference histogram, a frequency histogram, a saturation histogram, a maximum luminance of the image, a minimum luminance of the image, an average luminance of the image, a value of a black area in the image, and a value of a white area in the image;
3. The image processing device according to claim 1 or 2 . An output unit that outputs the corrected image.
The image processing device according to claim 1 . The image processing device
Obtain the image and the trained model,
Extracting a plurality of types of feature amounts based on the image;
Detecting correction parameters, which are parameters for correcting image quality of the image, using the plurality of types of feature amounts and the trained model;
determining whether a difference between the detected correction parameter and a preset correction parameter is equal to or smaller than a preset threshold value;
If the difference is equal to or smaller than the threshold, a process for correcting the image quality of the image is executed using the detected correction parameters.
Image processing methods. The image processing device includes:
Obtain the image and the trained model,
Extracting a plurality of types of feature amounts based on the image;
Detecting correction parameters, which are parameters for correcting image quality of the image, using the plurality of types of feature amounts and the trained model;
determining whether a difference between the detected correction parameter and a preset correction parameter is equal to or smaller than a preset threshold value;
If the difference is equal to or smaller than the threshold, a process for correcting the image quality of the image is executed using the detected correction parameters.
An image processing program that performs the processing.

Images