A white balance correction method based on white objects

Abstract

The invention discloses a white balance proofreading method based on white objects, which includes: using a large number of acquired white object samples to establish a white object contrast model through continuous training and learning; using the above white object contrast model to apply to on-site capture, and identifying images And extract the white object area in the image, calculate its pixel mean and color gain value, thereby adjusting the color to achieve white balance correction of the image; the invention has simple implementation principle, low resource cost, fast calculation time, and ideal and accurate white balance effect; the invention The tissue recognition technology used is based on the convolutional neural network algorithm, learns from large sample data, gradually extracts high-level features of the image, classifies the features, and completes the recognition. It can cope with a certain degree of offset, scale change and deformation of the tissue, ensuring The strong separability of features has an ideal recognition effect for feature classification, reducing the dependence of recognition on external conditions and reducing the complexity of the model.

Description

technical field [0001] The invention relates to the field of image color processing, in particular to a white balance correction method based on a white object. Background technique [0002] Traditional Chinese medicine includes the four inspections of looking, smelling, asking, and feeling, and tongue diagnosis is the key content of inspection. In the theory of traditional Chinese medicine, no matter how complicated the pathological symptoms of the internal organs of the human body are, the nature of the disease, the depth of the disease, and the state of qi and blood can be judged intuitively and quickly by observing the tongue image. It is a simple and effective medical treatment. Auxiliary diagnosis and differential methods. In traditional inspection, doctors generally observe the shape and color of the patient's tongue with the naked eye, and then diagnose the disease based on medical experience. Due to the large population and serious aging problem in our country, th...

AI Technical Summary

Problems solved by technology

Since the collected images rely on terminal shooting equipment, such as mobile phone cameras, this equipment does not have the adaptability of human eyes, and when shooting under different light conditions, color reproduction distortion occurs due to the imbalance of CCD output; therefore, the above tongue diagnosis There are often chromatic aberration problems in the collected images of products, which leads to deviations in medical diagnosis, and it is impossible to achieve targeted treatment and prevention of patients' diseases

Images