System and method for detecting dangerous substance by using artificial intelligence

1 . A hazardous substance detection system using artificial intelligence, the system comprising:

an X-ray output means for outputting X-rays to an inspection target object;

an X-ray detector configured to generate and output an X-ray image and an X-ray color resulting from the X-rays transmitting through the inspection target object;

database configured to store therein first matching of a component, a chemical formula, an atom with the largest atomic number, and an X-ray color with an object name, and second matching of a state, a component, a chemical formula, an atom with the largest atomic number and an X-ray color with a hazardous substance; and

a computer device electrically connected to the X-ray detector,

wherein the computer device is configured to:

execute deep learning on each of the first and second matchings using a deep learning algorithm; and

determine whether the inspection target object is or contains a hazardous substance, based on the X-ray image, the X-ray color, and the deep learning results,

wherein when the inspection target object contains sodium (Na) of an atomic number 11 and other atoms other than sodium (Na) in the periodic table, the X-ray detector is configured to display, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the sodium (Na),

wherein when the inspection target object contains potassium (K) of an atomic number 19 and other atoms other than potassium (K) in the periodic table, the X-ray detector is configured to display, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the potassium (K),

wherein when the inspection target object contains niobium (Nb) of an atomic number 41 and other atoms other than niobium (Nb) in the periodic table, the X-ray detector is configured to display, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the niobium (Nb),

wherein when the inspection target object is free of the sodium (Na), the potassium (K), and the niobium (Nb), the X-ray detector is configured to display, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among atoms contained in the object,

wherein the computer device includes a processor,

wherein the processor is configured to derive an element as the input information to the training engine from the input data including structured data and unstructured data,

wherein the processor is configured to:

convert the unstructured data including an image and a voice except for a text into text data;

extract information from the converted text data; and

derive the elements to be input to the training engine from the extracted information,

wherein processor is configured to

self-organize DNA mission using the elements derived by comparing and evaluating the elements input over time with elements within a mission of an organization based on each predefined group, and organizing and generating a DNA mission that changes over time,

self-construct deep learning-based artificial neural network DNA model using the self-organized DNA, mission by constructing the artificial neural network DNA model capable of performing the deep learning by itself by using the DNA mission, and

train the self-constructed DNA model,

wherein, based on a training result, processor is configured to perform situation understanding and scheduling, decision making and prediction, and recommendation and taking-action, and

wherein the processor is configured to:

understand a given situation or grasp an intention, and to provide scheduling to a decision maker based on a result of understanding the situation or understanding the intention;

identify the object name based on the X-ray image input from the X-ray detector, to predict whether the object is or contains a hazardous substance based on the X-ray color, and to provide a prediction result; and

recommend, based on an analysis result and prediction result, a decision for the given situation and to provide a taking-action according to the decision.

2 . The system of claim 1 , wherein the atom having the largest atomic number among atoms constituting each of an object of the object name and the hazardous substance is defined as a representative atom,

wherein the database is configured to store therein:

a first correspondence between a first group and first color data, wherein in the first group, the representative atom has the atomic number in a range of atomic numbers 1 to 10 in the periodic table;

a second correspondence between a second group and second color data, wherein in the second group, the representative atom has the atomic number in a range of atomic numbers 12 to 18 in the periodic table;

a third correspondence between a third group and third color data, wherein in the third group, the representative atom has the atomic number in a range of atomic numbers 20 to 40 in the periodic table; and

a fourth correspondence between a fourth group and fourth color data, wherein in the fourth group, the representative atom has the atomic number in a range of atomic numbers 42 to 118 in the periodic table.

3 . The system of claim 2 , wherein the X-ray detector is configured to display, as the X-ray color:

the first color data when the inspection target object belongs to the first group;

the second color data when the inspection target object belongs to the second group;

the third color data when the inspection target object belongs to the third group; and

the fourth color data when the inspection target object belongs to the fourth group.

4 . The system of claim 3 , wherein the first color data includes an orange-based color,

the second color data includes a green-based color,

the third color data includes a blue-based color, and

the fourth color data includes a black-based color.

5 . The system of claim 1 , wherein the inspection target object is a glass bottle containing nitromethane therein,

wherein the X-ray output means outputs the X-rays to the glass bottle,

wherein the X-ray detector generates an X-ray image including a shape of the glass bottle and outputs the X-ray color,

wherein the X-ray color includes:

a green color corresponding to an atom (Si) having the largest atomic number among atoms constituting the glass bottle; and

an orange color corresponding to an atom (O) having the largest atomic number among atoms constituting nitromethane (CH 3 NO 2 ) received in the glass bottle, wherein the computer device is configured to:

determine that the inspection target object is the glass bottle based on the X-ray image, based on the deep learning results;

determine that the inspection target object contains the hazardous substance based on the X-ray color including the orange color; and

output a warning message and/or a warning sound.

6 . The system of claim 1 , wherein the inspection target object is a glass bottle containing a radioactive substance therein,

wherein the X-ray output means outputs the X-rays to the glass bottle,

wherein the X-ray detector generates an X-ray image including a shape of the glass bottle and outputs the X-ray color,

wherein the X-ray color includes:

a green color corresponding to an atom (Si) having the largest atomic number among atoms constituting the glass bottle; and

a black color corresponding to an atom having the largest atomic number among atoms constituting the radioactive substance received in the glass bottle, wherein the computer device is configured to:

determine that the inspection target object is the glass bottle based on the X-ray image, based on the deep learning results;

determine that the inspection target object contains the hazardous substance based on the X-ray color including the black color; and

output a warning message and/or a warning sound.

7 . A hazardous substance detection method using artificial intelligence based on first matching of a component, a chemical formula, an atom with the largest atomic number, and an X-ray color with an object name, and second matching of a state, a component, a chemical formula, an atom with the largest atomic number and an X-ray color with a hazardous substance,

wherein the method comprises:

(a) executing, by a computer device, deep learning on the first matching using a deep learning algorithm;

(b) executing, by the computer device, deep learning on the second matching using the deep learning algorithm;

(c) outputting, by X-ray output means, X-rays to an inspection target object;

(d) generating and outputting, by an X-ray detector, an X-ray image and an X-ray color resulting from the X-rays transmitting through the inspection target object; and

(e) determining whether the inspection target object is or contains a hazardous substance, based on the X-ray image, the X-ray color, and the deep learning results,

wherein in the (d), when the inspection target object contains sodium (Na) of an atomic number 11 and other atoms other than sodium (Na) in the periodic table, the X-ray detector displays, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the sodium (Na),

wherein in the (d), when the inspection target object contains potassium (K) of an atomic number 19 and other atoms other than potassium (K) in the periodic table, the X-ray detector displays, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the potassium (K),

wherein in the (d), when the inspection target object contains niobium (Nb) of an atomic number 41 and other atoms other than niobium (Nb) in the periodic table, the X-ray detector displays, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among the atoms other than the niobium (Nb),

wherein in the (d), when the inspection target object is free of the sodium (Na), the potassium (K), and the niobium (Nb), the X-ray detector displays, as the X-ray color, a color corresponding to a group including an atom having the largest atomic number among atoms contained in the object,

wherein the computer device includes a processor,

wherein the processor is configured to derive an element as the input information to the training engine from the input data including structured data and unstructured data,

wherein the processor is configured to:

convert the unstructured data including an image and a voice except for a text into text data;

extract information from the converted text data; and

derive the elements to be input to the training engine from the extracted information,

wherein the processor is configured to

self-organize DNA mission using the elements derived by comparing and evaluating the elements input over time with elements within a mission of an organization based on each predefined group, and organizing and generating a DNA mission that changes over time,

self-construct deep learning-based artificial neural network DNA model using the self-organized DNA, mission by constructing the artificial neural network DNA model capable of performing the deep learning by itself by using the DNA mission, and

train the self-constructed DNA model,

wherein, based on a training result, the processor is configured to perform situation understanding and scheduling, decision making and prediction, and recommendation and taking-action, and

wherein the processor is configured to:

understand a given situation or grasp an intention, and to provide scheduling to a decision maker based on a result of understanding the situation or understanding the intention;

identify the object name based on the X-ray image input from the X-ray detector, to predict whether the object is or contains a hazardous substance based on the X-ray color, and to provide a prediction result; and

recommend, based on an analysis result and prediction result, a decision for the given situation and to provide a taking-action according to the decision.

8 . The method of claim 7 , wherein the atom having the largest atomic number among atoms constituting each of an object of the object name and the hazardous substance is defined as a representative atom,

wherein in a first group, the representative atom has the atomic number in a range of atomic numbers 1 to 10 in the periodic table;

wherein in a second group, the representative atom has the atomic number in a range of atomic numbers 12 to 18 in the periodic table;

wherein in a third group, the representative atom has the atomic number in a range of atomic numbers 20 to 40 in the periodic table; and

wherein in a fourth group, the representative atom has the atomic number in a range of atomic numbers 42 to 118 in the periodic table,

wherein in the (d), the X-ray detector displays, as the X-ray color:

first color data when the inspection target object belongs to the first group;

second color data when the inspection target object belongs to the second group;

third color data when the inspection target object belongs to the third group; and

fourth color data when the inspection target object belongs to the fourth group.

9 . The method of claim 8 , wherein the first color data includes an orange-based color,

the second color data includes a green-based color,

the third color data includes a blue-based color, and

the fourth color data includes a black-based color.

10 . The method of claim 8 , wherein in the (c), the X-ray output means outputs the X-rays to a glass bottle as the inspection target object,

wherein in the (d), the X-ray detector generates an X-ray image including a shape of the glass bottle and outputs the X-ray color, wherein the X-ray color includes:

a green color corresponding to an atom (Si) having the largest atomic number among atoms constituting the glass bottle; and

an orange or black color corresponding to an atom having the largest atomic number among atoms constituting a substance received in the glass bottle,

wherein the (e) includes:

determining that the inspection target object is the glass bottle based on the X-ray image, based on the deep learning results;

determining that the inspection target object contains the hazardous substance based on the X-ray color including the orange or black color; and

outputting a warning message and/or a warning sound.