TWI618027B - 3d hand gesture image recognition method and system thereof with ga - Google Patents

Abstract

A genetic algorithm three-dimensional gesture image recognition system comprises a light field photography unit, a calculation unit and an output unit. A light gesture is taken by the light field photographing unit to obtain a 3D gesture image. The calculating unit is connected to the light field photographing unit, and then projecting the 3D gesture image to a predetermined recognition space to obtain a feature vector, and processing the feature vector by using a genetic algorithm to obtain an adaptive feature vector, and then The adaptive feature vector and the plurality of sample feature vectors are compared for classification to classify the 3D gesture image to identify the type of the 3D gesture image. The output unit is connected to the calculation unit to output the type of the 3D gesture image.

Description

Gene calculus three-dimensional gesture image recognition method and system thereof

The invention relates to a method and system for recognizing a three-dimensional gesture image of a gene algorithm; in particular, a method for recognizing a three-dimensional gesture image of a 2D optimal principal component analysis combined with a genetic algorithm (GA) And its system.

A related application device for recognizing gesture image recognition, for example, the Republic of China Patent Publication No. M382675, a new type of surveillance camera control device based on gesture recognition, which discloses that a camera lens can be turned up and down, left and right, and zoomed in. And remote control and other action command control devices. Using the gesture image of the user captured by the gesture camera to recognize the movement of the user's gestures up, down, left, right, front, and back, and to send the camera up, down, left, and right to the surveillance camera. To zoom in and out, without having to manipulate the mouse or joystick to release the above control message.

The aforementioned M382675 first uses a gesture camera to capture the user's hand image, and then uses the gesture displacement detection module to detect and recognize the user's hand position and trend, thereby calculating the direction of the camera to be controlled, and then transmitting The signal control monitors the lens of the camera to perform actions of going up, down, left, right, zooming in and out. The gesture displacement detecting module first calculates the position and area of the hand in the gesture camera, and serves as the origin of the gesture recognition. This part can also be used in the gesture camera. The origin of a gesture recognition is preset on the screen as a displacement detection reference point for subsequent gesture recognition, and a threshold value of a minimum area of the gesture, a threshold value of a minimum moving distance of the gesture, and a minimum area change of a gesture are respectively defined. Threshold value.

In the foregoing No. M382675, after the size of the gesture area of the user is greater than a preset minimum threshold value of the gesture, the device starts to activate the gesture recognition function, and uses the gesture displacement detection module to detect the displacement of the current gesture position and the origin of the gesture. the amount. When the displacement of the gesture movement is greater than a preset threshold value, the gesture movement is considered to occur. Conversely, when the displacement amount of the gesture movement is less than the threshold value, no gesture movement occurs; and the gesture displacement detection is performed. The module also detects the current gesture position and the direction of displacement of the gesture origin. If the gesture moves up, down, left or right, a control signal is triggered to drive the camera lens to move up, down, left or right. In addition, the gesture displacement detection module uses image recognition technology to determine the movement of the gesture in the range of motion; when the gesture moves forward, the area thereof is larger than the area threshold, and a control signal is triggered to drive the surveillance camera lens to zoom in. When the gesture moves backwards, its area will be smaller than the area threshold, triggering a control signal to drive the camera lens to zoom out.

Another application device for recognizing gesture image recognition, for example, the gesture recognition system of the Republic of China Patent No. I298461 and its method and invention patent, which discloses a gesture recognition system and a method thereof applied to an image capture device Notebook computer. The user can directly align the image capture device with a preset gesture, and the notebook computer will execute the corresponding application or function option of the gesture action to increase the user execution application or function options. The convenience of time.

Another related application device for recognizing gesture image recognition, for example, the gesture recognition system of the Republic of China Patent No. I395145 and the method thereof, and the invention patent, which discloses that a gesture recognition system includes: a photographing device for obtaining a possible natural gesture Image; one processor used Find the skin contour of the skin color part from the image, and then classify the skin color contour into a plurality of contour fragments of different angles; an operation engine has a plurality of parallel operation units and a plurality of gesture template libraries of different angle categories, A plurality of parallel computing units respectively find a gesture template that is closest to the contour fragments in the gesture template library of different angle categories; an optimal template selection means, which is free from the plurality of approximate gesture templates found by the plurality of parallel operation units Then, an optimal gesture template is selected; and a display terminal is used to display an image of the best gesture template; thereby achieving the purpose of not recognizing the marker without using any marker.

Another conventional gesture image recognition related method and system, for example, the image-based motion gesture recognition method and system invention patent of the Republic of China Patent Publication No. I431538, discloses an image-based motion gesture recognition method and system. The method includes the steps of: receiving a plurality of image frames; performing a gesture detection according to the plurality of image images to obtain a first gesture, and if the first gesture conforms to a preset start gesture, according to the plurality of image frames The part position performs a movement tracking to obtain a movement gesture; during the execution of the movement tracking, performing gesture detection according to the plurality of image images to obtain a second gesture, if the second gesture meets a preset end Gesture, stop moving tracking.

Another conventional method and system for recognizing gesture image recognition, for example, the inventor proposes a gesture image recognition method for singular value decomposition processing complex background using the singular value decomposition method and a system patent of the invention, which discloses a singular value Decompose a gesture image recognition method that handles complex backgrounds. The method comprises: decomposing an original gesture image by using a singular value decomposition method to obtain a gain gesture image; removing a dark background from the gain gesture image to obtain at least one type of skin pixel block; and utilizing a skin color detection method The method performs skin color detection on the skin pixel block to remove the remaining background of the skin pixel block. The gesture image recognition system comprises an input unit, a calculation unit and an output unit. The input form The input unit is configured to input the original gesture image, and the calculation unit is configured to remove the dark background and the remaining background from the gain gesture image, and the output unit is configured to output a skin motion gesture image.

Another conventional method and system for recognizing gesture image recognition, for example, the inventor proposes a gesture image recognition method using image direction alignment and a system invention patent after the Republic of China Patent Publication No. I444908, which discloses an image orientation pair The gesture image recognition method being processed. The method includes: inputting a skin color gesture image; calculating a global heart in the skin color gesture image; selecting a region of interest in the skin color gesture image; selecting a sub region using the region of interest; and calculating a subregion in the subregion Regional nature; use this global nature and regional nature to calculate a pair of positive angles. The gesture image recognition system comprises an input unit, a calculation unit and an output unit. The input unit is configured to input the skin color gesture image, and the calculation unit selects the region of interest and the sub-region in the skin color gesture image to calculate the global nature heart and the region property core, and then calculate the alignment angle, and the output is The unit is used to output the pair of positive angles.

Another method and system for recognizing gesture image recognition, for example, the inventor proposes a gesture image recognition method using singular value decomposition for ray compensation processing and a system for inventing the invention, which discloses a singularity Value decomposition light compensation processing gesture image recognition method. The method comprises: inputting a gesture image; applying a singular value decomposition method to the gesture image; calculating at least one ray compensation coefficient by a ray compensation method; and performing ray compensation processing on the gesture image by using the ray compensation coefficient to obtain a gesture image Light compensated gesture image. The gesture image recognition system comprises an input unit, a calculation unit and an output unit. The input unit is configured to input the original gesture image, and the calculating unit calculates the light compensation coefficient by using the light compensation method, and performs light compensation processing on the gesture image by using the light compensation coefficient to obtain the light compensation gesture image, and the output is The unit is configured to output the light compensation gesture image.

In addition, the related application techniques of conventional gesture image recognition have been disclosed in some U.S. patents, for example, U.S. Patent No. 7,702,130, User interface apparatus using hand gesture recognition and method thereof, No. 7,680,295, Hand-gesture based interface apparatus〞 No. 6,215,890, Hand gesture recognizing device〞, No. 6,002,808, Hand gesture control system〞, and No. 5,594,469, Hand gesture machine control system〞. The foregoing Japanese patents and U.S. patents are only for the purpose of the present invention and are not intended to limit the scope of the present invention.

Although the aforementioned patents have disclosed related gesture image recognition techniques, they do not provide related techniques for three-dimensional gesture image recognition. In fact, in terms of gesture image recognition technology, it is necessary to simplify the complicated system architecture or omit complicated pre-processing procedures, otherwise it will affect the reliability of gesture image recognition. Therefore, the conventional gesture image recognition technology necessarily has the need to further provide three-dimensional gesture image recognition.

In view of the above, the present invention provides a genetic algorithm for three-dimensional gesture image recognition and a system thereof, which utilizes a light field photographing unit to take a gesture to obtain a 3D gesture image, and the 3D gesture image. Projecting to a predetermined recognition space to obtain a feature vector, and processing the feature vector by using a genetic algorithm to obtain an adaptive feature vector, and comparing the adaptive feature vector with the plurality of sample feature vectors. Gesture image recognition can be completed to improve the reliability of the conventional gesture image recognition technology.

The main object of the preferred embodiment of the present invention is to provide a method for recognizing a three-dimensional gesture image of a genetic algorithm and a system thereof, which utilizes a light field photographing unit to take a gesture motion to obtain a 3D gesture image, and project the 3D gesture image to a predetermined recognition space to obtain a feature vector, The eigenvector is processed by a genetic algorithm to obtain an adaptive eigenvector, and the adaptive eigenvector is compared with the plurality of sample eigenvectors to achieve the purpose of improving the reliability of the gesture image recognition.

In order to achieve the above object, a three-dimensional gesture image recognition method for a genetic algorithm according to a preferred embodiment of the present invention includes: capturing a gesture by using a light field photographing unit to obtain a 3D gesture image; and projecting the 3D gesture image to a predetermined recognition space. Obtaining a feature vector; processing the feature vector by using a gene algorithm to obtain an adaptive feature vector; and classifying the adaptive feature vector and the plurality of sample feature vectors to classify the 3D gesture image In order to identify the type of the 3D gesture image.

In the preferred embodiment of the present invention, the 3D gesture image includes a plane image information and a depth information.

In the preferred embodiment of the present invention, the 3D gesture image is a 3D gesture contour image or a 3D gesture solid image.

In a preferred embodiment of the present invention, the 3D gesture image is converted into a projected color space by using a projected color space method to obtain an R channel image information, a G channel image information, and a B channel image information.

In the preferred embodiment of the present invention, when the 3D gesture image is projected to the predetermined recognition space, projection is performed by principal component analysis or two-dimensional optimal principal component analysis.

In the preferred embodiment of the present invention, an adaptive function or a shell-like approximate function is used before the feature vector is processed by the genetic algorithm.

In the preferred embodiment of the present invention, when the adaptive feature vector and the plurality of sample feature vectors are compared for classification, the nearest neighbor method is used for comparison and classification.

The preferred embodiment of the present invention processes the 3D gesture image using a horse distance function or a horse cosine distance function before classifying the adaptive feature vector.

In order to achieve the above object, a genetic algorithm three-dimensional gesture image recognition system according to a preferred embodiment of the present invention comprises: a light field photographing unit that takes a gesture motion to obtain a 3D gesture image; and an arithmetic unit connected to the light field The photographing unit projects the 3D gesture image into a predetermined recognition space to obtain a feature vector, and processes the feature vector by using a genetic algorithm to obtain an adaptive feature vector, and then the adaptive feature vector and the plurality of The sample feature vectors are categorized to classify the 3D gesture image to identify the type of the 3D gesture image; and an output unit coupled to the calculation unit to output the type of the 3D gesture image.

In the preferred embodiment of the present invention, the 3D gesture image is a 3D gesture contour image or a 3D gesture solid image.

In a preferred embodiment of the present invention, the 3D gesture image is converted into a projected color space by using a projected color space method to obtain an R channel image information, a G channel image information, and a B channel image information.

In the preferred embodiment of the present invention, when the 3D gesture image is projected into the predetermined recognition space, the principal component analysis method or the two-dimensional optimal principal component analysis method is used for projection, and the adaptive feature vector and the plurality of sample features are used. When the vectors are compared for classification, the nearest neighbor method is used for comparison classification.

S1‧‧‧ steps

S2‧‧‧ steps

S3‧‧‧ steps

S4‧‧‧ steps

10‧‧‧Light field photography unit

20‧‧‧ calculus unit

30‧‧‧Output unit

FIG. 1 is a flow chart showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention.

Figure 2: Image analysis of three-dimensional gestures of genetic algorithm in a preferred embodiment of the present invention A block diagram of the system.

3A and 3B are diagrams showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention and a system for capturing nine 3D gesture contour images and nine solid images of 3D gestures.

FIG. 4 is a schematic diagram showing the relationship between the recognition rate and the feature vector after the 3D gesture contour image and the 3D gesture solid image are recognized by the k =3 condition in the preferred embodiment of the present invention.

5A and 5B are diagrams showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for taking nine kinds of 3D gesture contour images and nine solid images of 3D gestures for each of the three rotation variation conditions.

6(A) and 6(B) are diagrams showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for generating a PCS projection image from an original 3D gesture through an RGB image and processing thereof.

7(a) to 7(d): a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention and a system thereof, which are tested by adding various Gaussian noise variations conditions, and then the original 3D gesture contour image and a series of images A comparison diagram for comparing 3D gesture contour images.

8(a) to 8(d): a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for adding a variety of Gaussian noise variations to test a solid image of the original 3D gesture with a series of Test a comparison of solid images of 3D gestures.

In order to fully understand the present invention, the preferred embodiments of the present invention are described in detail below, and are not intended to limit the invention.

The gene calculus three-dimensional gesture image recognition method and system thereof according to the preferred embodiment of the present invention are applicable to various gesture recognition devices and related application devices, for example, various types of computer systems, home appliance product control systems (such as objects) Networking], automated control systems, medical care systems, or security systems, but are not intended to limit the scope of the invention.

FIG. 1 is a flow chart showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention. 2 is a block diagram showing a three-dimensional gesture image recognition system for genetic algorithm according to a preferred embodiment of the present invention, which corresponds to FIG. Referring to FIGS. 1 and 2, for example, a three-dimensional gesture image recognition system according to a preferred embodiment of the present invention includes a light field capturing unit 10, a calculation unit 20, and an output unit 30. The calculation unit 20 is suitably connected to the light field photographing unit 10, and the output unit 30 is suitably connected to the calculation unit 20.

Referring to FIGS. 1 and 2, for example, the light field photographing unit 10 includes a plurality of micro lenses, and the microlens lens forms a microlens array, and the microlens array is used for Actively take a depth of field image. In addition, the depth information of the depth image of the field is calculated by using a plurality of depth calculation parameters. The depth calculation parameter includes a maximum virtual depth, an approximate resolution level, a pixel step, a minimum standard deviation, a minimum correlation, and a curvature pole. Extremum curvature, correlation patch diameter, correlation patch stride, depth fill, and depth post-process.

Referring to FIG. 1 and FIG. 2 again, the three-dimensional gesture image recognition method according to the preferred embodiment of the present invention includes the step S1: first, using the light field photographing unit 10 or the light field image input unit having a similar function at a predetermined distance. [Example: 50 to 70 cm] Take a gesture to get a 3D gesture image (for example: 100x100 pixels). For example, the 3D gesture image includes a plane image information and a depth information. The 3D gesture The image is a 3D gesture contour image or a 3D gesture solid image.

FIG. 3A is a schematic diagram of a three-dimensional gesture image recognition method for a gene calculus according to a preferred embodiment of the present invention, and a schematic diagram of a system for capturing nine 3D gesture contour images (a planar rotation gesture contour image, IC), which includes a zero-to-eight 3D gesture gray scale. [grayscale] contour image for use in in-plane rotation variation. In contrast, FIG. 3B discloses a schematic diagram of a three-dimensional gesture image recognition method for a genetic algorithm according to a preferred embodiment of the present invention, and a schematic diagram of a system for capturing nine solid images of a 3D gesture (solid image of a planar rotation gesture, IS), which includes zero to eight 3D. Gesture grayscale solid image for use in planar rotation variations.

Referring to FIG. 1 and FIG. 2 again, the three-dimensional gesture image recognition method according to the preferred embodiment of the present invention includes the step S2: then, the 3D gesture image is projected to a predetermined recognition space to obtain a feature vector. For example, the preferred embodiment of the present invention projects the features of the 3D gesture image into a relatively easily discernible space using a principal component analysis (PCA) or a 2D optimal PCA method.

The principal component analysis can be applied to the dimensionality reduction of high-dimensional data, and the degree of variation of the data is preserved, because the one-dimensional PCA needs to straighten the dimensions of the input training image. Taking the image of ( m × n ) size as an example, when the covariation matrix is calculated, a matrix of ( m × n ) × ( m × n ) is generated, which requires a great calculation time for the step of calculating the feature vector. . Therefore, the original covariation matrix is changed to the following formula:

Where C _i is the changed covariation matrix, L is the number of training samples, and X _tr is the training image. The average of all training images. In this way, the covariation matrix dimension is reduced to L × L and the time required to calculate the projection substrate is greatly reduced. The preferred embodiment of the present invention uses the singular value decomposition [SVD] to operate the covariation matrix as follows:

The formula obtained eigenvalue matrix Σ _i resulting eigenvalue matrix [Sigma same original covariance matrix C through singular value decomposition, and the eigenvector matrix shall be to (X _tr - ) * U _i square forms the eigenvector matrix U . Then, the original training data is projected into the PCA space by using the feature vector matrix to obtain the PCA processed training sample feature vector F _tr .

Another preferred embodiment of the present invention utilizes two-dimensional optimal principal component analysis to select k feature vectors from n total feature vectors [feature vector set U = {1, 2, ..., n }]. The eigenvector set U = {1, 2, ..., n } includes a first subset [ X ] and a second subset [ Y ], wherein the first subset [ X ] is a subset of the selected feature vectors, and the second The subset [ Y ] is a subset of the remaining feature vectors. Preferred embodiment of the present invention to employ operators φ is determined from the X minimum feature subset selection least discriminant feature subset [] and ψ operators to select the maximum is determined from the [Y] most discriminant feature subset, thus | Ω | = κ , , The preferred embodiment of the present invention employs the following functions: , κ D

J (Ω) is used to evaluate the subset Ω by leave-one-out.

Referring again to Figures 1 and 2, a preferred embodiment of the present invention The three-dimensional gesture image recognition method comprises the step S3: Next, the feature vector is processed by a genetic algorithm [GA] to obtain an adaptive feature vector. For example, when the feature vector is processed by the genetic algorithm, a fitness function or a Bayesian likelihood function is used.

The preferred embodiment of the present invention adopts a Bayesian approximate function as follows:

κ/n is the ratio of the number of eigenvectors to the total number of eigenvectors, η ₁ , η ₂ , η ₃ , η ₄ are the action coefficients, TPR is the true positive rate, and TNR is the true negative rate. Rate], FNR is false negative rate, FPR is false negative rate [false posi

The preferred embodiment of the present invention inputs the 3D gesture image [matrix a R ^ζ ,ζ n 〕 a column (column per image) to obtain a two-dimensional optimal principal component analysis, where ζ is the size of the image; then, using the genetic algorithm [GA] to select κ two-dimensional optimal principal components, so that for each Training subject vector _{c ml} produces a projection vector [ a projection vector ] _{a ml} , and the operation of the projection vector is as follows: , m =1,2,..., M , l =1,2,..., L , , κ < n

Wherein a _ml is the main vector of the m l of images, The two-dimensional optimal principal component of the κ corresponds to the m -th main vector selected from the co-variance matrix of the sample.

Referring to FIG. 1 and FIG. 2 again, the three-dimensional gesture image recognition method according to the preferred embodiment of the present invention includes the step S4: then, the adaptive feature vector and the plurality of sample feature vectors are compared and classified to classify the 3D gesture image to identify the type of the 3D gesture image. For example, when the adaptive feature vector is compared with a plurality of sample feature vectors for comparison, the nearest neighbor method [ KNN , k- nearest neighbors] is used for comparison classification. The k training samples closest to the test image in the feature space are classified, the test image is projected into the PCA space, and compared with the training samples, so that the similarity is calculated as follows:

Where S _k is the calculated similarity matrix, k is the number of the nearest training samples, N is the maximum value of the feature roots, F _te and F _tr are the feature vectors of the test and training samples, and the similarity is selected according to the k value. The nearest k training samples determine which of the selected training samples is closest to the classification of the selected training samples for classification prediction.

In a preferred embodiment of the present invention, the 3D gesture image is processed using a Mahalanobis distance function or a cosine distance function before classifying the adaptive feature vector. The horse distance function of the preferred embodiment of the present invention is calculated as follows:

Where c _t is the projection vector of the two-dimensional optimal principal component of the t- test, ζ _mt is the distance from the m- category to the t- test image, and the mean u _m and the variation value v _m are selected from the m- th category. The optimal principal component of the dimension is verified by the leave-one-out method. γ is displayed as the category identification code, and δ is displayed as the threshold value.

FIG. 4 is a schematic diagram showing the relationship between the recognition rate and the feature vector after the 3D gesture contour image and the 3D gesture solid image are recognized by the k =3 condition in the preferred embodiment of the present invention. Referring to FIG. 4, the preferred embodiment of the present invention uses a recognition 3D gesture image including a plane rotation gesture contour image (IC), a plane rotation gesture solid image [IS], a depth rotation gesture contour image [OC], and a deep rotation gesture. Solid image [OSG] and projected color space depth rotation gesture solid image [OSG PCS].

FIG. 5A is a schematic diagram of a three-dimensional gesture image recognition method for a genetic algorithm according to a preferred embodiment of the present invention, and a system for capturing nine kinds of 3D gesture contour images (plane rotation gesture contour image, OC) by using three plane rotation variation conditions per image. In contrast, FIG. 5B discloses a three-dimensional gesture image recognition method for a genetic algorithm according to a preferred embodiment of the present invention, and a system for taking nine kinds of 3D gesture solid images by using three out-of-plane rotation conditions for each image. Deep rotation gesture solid image, OSG] schematic. For example, the preferred embodiment of the present invention uses rotation angles of 0 degrees, ±15 degrees, and ±30 degrees, respectively (three rotation angle gestures are arranged from left to right).

6(A) and 6(B) are diagrams showing a method for recognizing a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for generating a PCS projection image from an original 3D gesture through an RGB image and processing thereof. The genetic algorithm three-dimensional gesture image recognition method according to the preferred embodiment of the present invention will select at least one deep-rotation color gesture solid image, as shown in FIG. 6(A). Then, the depth-rotating color gesture solid image is used to obtain RGB three-channel image information by a projection color space method. When projecting the color space, first generate an R channel image, a G channel image and a B channel image, and then project the RGB three channel image information into the same space to obtain a PCS gesture solid projection image, such as the sixth ( B) The picture shows.

7(a) to 7(d) show a method for identifying a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for adding three Gaussian noise variations conditions [adding 10%, 20%, 30) % noise] After testing, the original 3D gesture contour image and a series of tests A schematic diagram of the comparison of 3D gesture contour images. Referring to Figures 7(a) through 7(d), the average of the Gaussian noise added to the original 3D gesture contour image (as shown in Fig. 7(a)) is 0, and the 3D gesture is The contour image is identified by adding 10%, 20%, and 30% noise, as shown in Figures 7(b) through 7(d). Table 1 shows that the variation of the 10%, 20%, and 30% noise is added to the 3D gesture contour image, and the TPR and TNR recognition rates are still as high as 100%.

8(a) to 8(d) show a method for identifying a three-dimensional gesture image of a genetic algorithm according to a preferred embodiment of the present invention, and a system for adding three Gaussian noise variations conditions (10%, 20%, 30%) for testing. A comparison of the original 3D gesture solid image with a series of test 3D gesture solid images. Referring to Figures 8(a) to 8(d), the average of the Gaussian noise added to the original 3D gesture solid image (as shown in Fig. 8(a)) is 0, and the 3D gesture is Solid images were identified by adding 10%, 20%, and 30% noise, as shown in Figures 8(b) through 8(d). Table 2 shows that the 10%, 20%, and 30% noise is added to the solid image of the 3D gesture, and the TPR and TNR recognition rates are still as high as 100%.

Table 2: Identification results of adding solid image of Gaussian noise 3D gesture

The foregoing preferred embodiments are merely illustrative of the invention and the technical features thereof, and the techniques of the embodiments can be carried out with various substantial equivalent modifications and/or alternatives; therefore, the scope of the invention is subject to the appended claims. The scope defined by the scope shall prevail. The copyright limitation of this case is used for the purpose of patent application in the Republic of China.

Claims (10)

A method for recognizing a three-dimensional gesture image of a genetic algorithm, comprising: capturing a gesture by using a light field photographing unit to obtain a 3D gesture image, wherein the 3D gesture image is a 3D light field gesture image, and the 3D light field gesture image Include a 3D light field information; project the 3D light field gesture image into a predetermined recognition space, and the predetermined identification space is a space to obtain a 3D light field feature vector; and process the 3D light field feature by using a genetic algorithm Obtaining an adaptive 3D light field feature vector; and classifying the adaptive 3D light field feature vector and the plurality of sample feature vectors to classify the 3D gesture image to identify the type of the 3D gesture image . The method for recognizing a three-dimensional gesture image according to the first aspect of the patent application scope, wherein the 3D light field gesture image comprises a plane image information and a depth information. The method for recognizing a three-dimensional gesture image according to the first aspect of the patent application scope, wherein the 3D light field gesture image is a 3D light field gesture contour image or a 3D light field gesture solid image. According to the third aspect of the patent application scope, the three-dimensional gesture image recognition method is performed, wherein the 3D light field gesture image is converted into a projection color space by using a projection color space method to obtain an R channel image information and a G channel image information. And a B channel image information. The method for recognizing a three-dimensional gesture image according to the first aspect of the patent application scope, wherein when the 3D light field gesture image is projected to the predetermined recognition space, the principal component analysis method or the two-dimensional optimal principal component analysis method is used. projection. The three-dimensional gesture image recognition method according to the first aspect of the patent application scope, wherein when the adaptive 3D light field feature vector and the plurality of sample feature vectors are compared for classification, the nearest neighbor method is used for comparison and classification. . A gene calculus three-dimensional gesture image recognition system includes: a light field photographing unit that takes a gesture motion, and the 3D gesture image is a 3D light field gesture image, and the 3D light field gesture image includes a 3D light field information Obtaining a 3D gesture image; a calculation unit connected to the light field photography unit, projecting the 3D light field gesture image to a predetermined recognition space, and the predetermined recognition space is a space to obtain a 3D light field The feature vector is processed by a gene algorithm to obtain an adaptive 3D light field feature vector, and the adaptive 3D light field feature vector and the plurality of sample feature vectors are compared and classified. The 3D light field gesture image is classified to identify the type of the 3D light field gesture image; and an output unit is connected to the calculation unit to output the type of the 3D light field gesture image. The three-dimensional gesture image recognition system according to the seventh aspect of the patent application scope, wherein the 3D light field gesture image is a 3D light field gesture contour image or a 3D light field gesture solid image. According to the genetic algorithm of claim 8, the three-dimensional gesture image recognition system, wherein the 3D light field gesture image is converted into a projection color space by using a projection color space method to obtain an R channel image information and a G channel image information. And a B channel image information. The three-dimensional gesture image recognition system according to the seventh aspect of the patent application scope, wherein when the 3D light field gesture image is projected to the predetermined recognition space, the principal component analysis method or the two-dimensional optimal principal component analysis method is used. Projection, and when the adaptive 3D light field feature vector is compared with a plurality of sample feature vectors for comparison, the nearest neighbor method is used for comparison classification.