AI-based apparatus and method for detecting a defect of a product

1 . An AI-based apparatus for detecting a defect of a product, comprising:

a sensor unit;

a detection unit;

a preprocessor configured to:

photograph a product to generate image data and measure at least one of a color, a saturation, a brightness, a transparency, and a reflectance of the product; and

detect a defect on the product by inputting the image data to a convolutional neural network (CNN) trained to detect a defect on a product surface;

convert at least one of the color, the saturation, and the brightness of the image data with respect to at least one of the color, the saturation, and the brightness of the product used as training data of the convolutional neural network;

wherein a number of convolution layers of the convolutional neural network is adjusted based on a defect detection difficulty determined according to at least one of the color, the saturation, the brightness, the transparency, and the reflectance of the product and a defect type, and

wherein the convolutional neural network performs a learning by receiving a predetermined size of training image data for a first defect type to divide the training image data into grids having a predetermined size, wherein the training image data is configured by a three-channel image obtained by dividing an image for the same product with respect to RGB and a size of the training image data is 448×448, and a size of the grid is 7×7; and

wherein the convolutional neural network performs the learning as many as the number obtained by dividing a size of one training image data by a size of the grid.

2 . The AI-based apparatus for detecting a defect of a product according to claim 1 , wherein the convolutional neural network learns N defect types.

3 . The AI-based apparatus for detecting a defect of a product according to claim 1 , wherein the number of convolution layers of the convolutional neural network is determined further based on a similarity between at least one of the color, the saturation, the brightness, the transparency, and the reflectance of the product used as the training data of the convolutional neural network and at least one of the color, the saturation, the brightness, the transparency, and the reflectance of the product.

4 . The AI-based apparatus for detecting a defect of a product according to claim 3 , wherein the number of convolution layers of the convolutional neural network is increased in inverse proportion to the similarity.

5 . The AI-based apparatus for detecting a defect of a product according to claim 1 , further comprising: a preprocessor which converts the image data based on at least one of the color, the saturation, the brightness, the transparency, and the reflectance of the product to input the converted data to the detection unit.

6 . The AI-based apparatus for detecting a defect of a product according to claim 5 , wherein the preprocessor performs auto cropping to extract a shape of the product from the image data to extract at least one feature of a brightness and a shadow of the product based on a shape of the automatically cropped product.

7 . The AI-based apparatus for detecting a defect of a product according to claim 5 wherein the preprocessor determines a frequency of an image sharpening filter based on the transparency of the product and converts the image data by applying the image sharpening filter to the image data.

8 . The AI-based apparatus for detecting a defect of a product according to claim 1 , wherein the detection unit detects a position of the defect, a size of the defect, and a type of the defect on the product and if a predetermined number or more of defects of the same position, same size, and same type occur in a predetermined consistency level, it is determined that the defect is not a defect, and wherein the detecting unit displays a bounding box on the image data based on the position of the defect and the size of the defect present on the product and outputs the image data displayed with the bounding box and the type of the defect.

9 . An AI-based method for detecting a defect of a product, comprising:

generating image data by photographing a product; measuring at least one of a color, a saturation, a brightness, a transparency, and a reflectance of the product; and

detecting a defect on the product by inputting the image data to a convolutional neural network trained to detect the defect on a product surface,

converting at least one of the color, the saturation, and the brightness of the image data with respect to at least one of the color, the saturation, and the brightness of the product used as training data of the convolutional neural network;

wherein a number of convolution layers of the convolutional neural network is adjusted based on a defect detection difficulty determined according to at least one of a color, a saturation, a brightness, a transparency, and a reflectance of the product and a defect type,

wherein the convolutional neural network performs a learning by receiving a predetermined size of training image data for a first defect type to divide the training image data into grids having a predetermined size, wherein the training image data is configured by a three-channel image obtained by dividing an image for the same product with respect to RGB and a size of the training image data is 448×448, and a size of the grid is 7×7; and

wherein the convolutional neural network performs the learning as many as the number obtained by dividing a size of one training image data by a size of the grid.

10 . The AI-based method for detecting a defect of a product, according to claim 9 , wherein the convolutional neural network learns N defect types.