Facial form prediction method based on gene data and generative adversarial convolutional neural network

Abstract

The invention discloses a facial form prediction method based on gene data and a generative adversarial convolutional neural network. The method comprises the following steps: S1, carrying out the data preprocessing of a human skull CT scanning image, and making a data set through combining with the corresponding gene data; S2, building a convolutional neural network; S3, training the convolutional neural network built in step S2 by using a training sample, detecting a detection sample by using the model every time the convolutional neural network is trained for four rounds, and judging a generated prediction surface type; and S4, selecting the round with the best effect in step S3 as a final result. The deep learning technology is utilized, the features are automatically extracted throughthe convolutional neural network, the complete facial form corresponding to the gene can be calculated in a very short time only by inputting the gene sequence into the model, and end-to-end prediction is achieved. While the prediction efficiency is improved, the considerable prediction accuracy is ensured.

Description

technical field [0001] The invention belongs to the field of genetic medicine, and relates to a method for predicting facial shape based on genetic data in combination with computer technology. Background technique [0002] Predicting the facial shape of the human body through genes has always been a research hotspot in the fields of criminal investigation and forensic medicine. However, traditional algorithms have limited computing power, and it is difficult to fully explore the complex correlation between genetic data and facial shape. [0003] At present, the traditional method of predicting facial shape based on genes mainly uses some algorithms of machine learning, and there are a lot of manual interactive work, such as manual labeling of a large amount of data, manual data cleaning, etc., not only the processing process is complicated, but also can be extracted. The feature information is also very limited. Contents of the invention [0004] In order to solve the ab...

AI Technical Summary

Problems solved by technology

[0002] Predicting the facial shape of the human body through genes has always been a research hotspot in the fields of criminal investigation and forensic medicine. However, traditional algorithms have limited computing power, and it is difficult to fully explore the complex correlation between genetic data and facial shape.

[0003] At present, the traditional method of predicting facial shape based on genes mainly uses some algorithms of machine learning, and there are a lot of manual interactive work, such as manual labeling of a large amount of data, manual data cleaning, etc., not only the processing process is complicated, but also can be extracted. The characteristic information of

Images