112results about How to "Conform to visual characteristics" patented technology

The invention discloses an objective evaluation method of no-reference video quality weighted by key frame image quality. The method comprises the steps of: firstly, preliminarily selecting a key frame according to a movement intensity method weighed by a human eye interest, then dividing the preliminarily selected key frame into a scene switching type key frame and a non-scene switching type key frame by correlation analysis; evaluating image quality of the non-scene switching type key frame, subdividing the non-scene switching type key frame into a content change type key frame and a quality change type key frame according to an evaluation result; finally carrying out weighted summation calculation on single frame quality by using a duration factor and a deterioration frequency factor of the quality change type key frame, and obtaining the quality evaluation result of an entire video sequence. According to the objective evaluation method, the defect that then existing evaluating technology cannot be coincided with an actual subjective feeling; computation complexity can be reduced on the premise of ensuring the evaluating performance; the objective evaluation method is suitable for automatic evaluation of various video applications of an existing network, simple in operating steps and easy to integrate.

The invention provides an infrared image and visible image fusion method based on guide filtering. The method comprises the steps of: firstly, expanding the guide filtering to multiple scales, and performing multi-scale decomposition on an infrared image and a visible image by using the multi-scale guide filtering, so as to obtain a low-frequency subband and a high-frequency subband; performing nonsubsampled directional filtering on the high-frequency subband, so as to obtain a directional subband coefficient; applying different fusion rules to the directional subband coefficient and a low-frequency subband coefficient, so as to obtain the corresponding directional subband coefficient and the low-frequency subband coefficient after fusion, and finally, performing directional filtering reconstruction and multi-scale reverse transformation based on the guide filtering, so as to obtain a final fusion image. According to a fusion result obtained by adopting the method, the edge and thermal radiation characteristics of an infrared image object are preferably kept, scene details of the visible image can be preferably reserved, and the information content of a fusion image is increased, so that the edge and details of the fusion image are enriched.

The invention discloses a video area-of-interest exacting method based on depth; the invention has the advantages of associatively utilizing texture video frame and depth video frame corresponding to the texture video frame, both of the texture video frame and the depth video frame are synchronous on time; the method comprises the steps as follows: firstly obtaining motion region of the texture video frame and binary mask image of contour texture, and each depth layer of the depth video frame and binary mask image of depth discontinuous regions, then obtaining the binary mask image of each depth object layer of each depth layer, and the binary mask image of contour feature region of each depth layer, then taking the binary mask image of contour feature region of each depth layer as feature, for scanning the binary mask image of each depth object layer, obtaining the area-of-interest in the binary mask image of each depth object layer, finally stacking the area-of-interest in the binary mask image of each depth object layer to obtain the area-of-interest of the texture video frame; the area-of-interest obtained by the method conforms to the semantic feature and has low calculating complexity and high precision.

The invention discloses an image nonlinearity enhancement method based on histogram subsection transformation. The method comprises the following steps of: S1, calculating probability density distribution of image gray levels; S2, calculating accumulated probability density functions of the gray levels; S3, dividing the gray levels of the images into three parts, wherein the first part of gray level range is between 0 and 85, the second part of gray level range is between 86 and 170, and the third part of gray level range is between 171 and 255; respectively calculating gray level transformation functions of the three parts of gray levels according to the accumulated probability density functions, namely proposing sectional gray level transformation functions; and S4, transforming the gray levels of the original images by using the sectional gray level transformation functions to acquire the distribution of new gray levels. By the method, the definition of the image is further improved, the excessive enhancement of the image is avoided, and the image vision effect meets the requirement of an eye vision system.

The invention discloses a low-illumination image enhancement method and system, and belongs to the technical field of image processing. The method comprises the steps of converting an input original image into an HIS space from an RGB space, keeping the saturability and the hue component of the image unchanged, enhancing the brightness component, converting the image with the brightness enhanced into the RGB space from the HIS space and outputting the image. According to the low-illumination image enhancement method and system, the image brightness is improved, extremely-dark areas (the areas close to black) are prevented from being changed into bright-colored areas after being enhanced, and the image adapts to the human visual characteristics better. Meanwhile, the complexity is low, the calculation amount is small, high-definition images can be processed in real time, and the performance requirement for hardware can be lowered.

The invention discloses a light-emitting control method, apparatus and system for an LED (light emitting diode) lamp. The light-emitting control method comprises the steps of receiving a control message; analyzing the control message; determining a control mode and a control target value of the LED lamp, wherein the control mode of the LED lamp comprises a lightening mode, a luminance regulation mode, an extinguishing mode and a toning mode, and the control target value comprises a three-color target duty ratio; determining a light-emitting gradual change parameter corresponding to the control mode of the LED lamp; calculating a gradual change PWM duty ratio corresponding to the light-emitting gradual change parameter in the toning process of the LED lamp, wherein the value range of the light-emitting gradual change parameter is from 0.005 to 0.05; outputting a PWM (pulse width modulation) signal to an actuator, wherein the duty ratio of the PWM signal is the gradual change PWM duty ratio; and controlling the LED lamp to give out light in a gradual change manner, wherein the frequency of the PWM signal is greater than 100 hertz. According to the light-emitting control method, apparatus and system for the LED lamp, the relatively soft and uniform light emission of the LED lamp can be realized, the emitted lights can better accord with visual characteristics of human body, and the user experience is improved.

The invention relates to a Retinex-theory-based low-illumination low-altitude remote sensing image enhancing method, mainly comprising the following steps of: transforming a low-illumination low-altitude remote sensing image from a red, green, blue (RGB) color space into a hue, saturation, intensity (HSI) color space; adjusting a luminance component I and a saturation component S on the premise of guaranteeing that a hue component H is not changed, so that the processed image accords with visual characteristics of eyes of a human body and a color distortion phenomenon is avoided; transforming the enhanced image from the HIS color space into the RGB color space; and performing convolution operation by using a laplace operator and an original image to obtain an edge image; and superposing the enhanced image and the edge image to obtain the finally enhanced image. By virtue of the method, the luminance and the contrast of the low-illumination low-altitude remote sensing image are improved, the color distortion phenomenon is avoided, and the low-illumination low-altitude remote sensing image has rich edge information and accords with the visual characteristics of the eyes of the human body.

The invention provides an image segmentation quality evaluating method based on an ROC Graph. The image segmentation quality evaluating method includes the following steps: firstly, dividing pixels in a to-be-evaluated segmentation image into four types by comparing with a reference segmentation image; secondly, leading in pixel space information, performing distance transform on pixels in the reference image, and calculating distance from each pixel to a target boundary; thirdly, giving a weight to a pixel at a corresponding position in the segmentation image according to distance coefficient of each pixel in the reference image; fourthly, utilizing the weights to calculate a weighted true positive rate wTPR and a weighted false positive rate wFPR, and depicting a point (wFPR, wTPR) in the ROC Graph; fifthly, according to distance from the point (wFPR, wTPR) to a point (0,1), scoring the segmentation image, wherein the higher the score is, the better the segmentation effect of the to-be-evaluated image is. By the image segmentation quality evaluating method, quality of the segmentation image can be evaluated effectively, and the evaluation result is more reasonable and more visual and consistent with subjective evaluation result of human vision.

The invention discloses a vehicle safety auxiliary video image processing method. Images acquired by a vehicle-mounted video device are processed. The method comprises the image enhancement, noise reduction and strong light inhibition method, the road line detection method and the obstacle monitoring method. According to the processing method of the invention, the improved Retinex algorithm is adopted to perform local enhancement on the images, and at the same time, the dynamic histogram mapping algorithm is adopted to perform overall image enhancement, the two enhancement results are combined in the weight form to obtain the final enhanced images, effects, such as image enhancement, noise elimination, color sharpening, etc., can be realized, the image contrast and definition can be improved to obtain images which are close to the original images and conform to human visual characteristics, and characteristics of road line detection and obstacle distance measurement can be realized.

The invention discloses a multi-focus noise image fusion method based on a cycle spinning-Contourlet domain hidden Markov tree model (CS-CHMT) and an improved dual-channel pulse-coupled neural network (IDPCNN). First, two multi-focus images containing a certain level of Gauss white noise are de-noised by use of the CS-CHMT model, and on the basis, a fusion strategy is designed by use of the IDPCNN to obtain a final fused image. According to the invention, by making use of the directional sensitivity of Contourlet transform height and the advantages of anisotropy, performing image de-noising by use of a hidden Markov tree (HMT) model, and introducing a cycle spinning technology to effectively suppress the pseudo Gibbs effect of the images near singular points, the PSNR value of de-noised images is improved. Compared with the traditional multi-focus image fusion method, the improved IDPCNN fusion method can effectively preserve more detailed information characterizing image features, greatly improve the quality of fused images and further improve the visual effect, and has real-time performance.

The invention provides a multi-modality image fusion method based on a region and human eye contrast sensitivity characteristic, comprising the following steps of: (1) respectively utilizing a non-subsampled Contourlet transform NSCT to carry out multi-scale decomposition on a source image to be fused to obtain each stage of sub-band coefficient of the source image; (2) respectively formulating a fusion rule of a low-frequency sub-band coefficient and each stage of high-frequency sub-band coefficient according to a human eye vision contrast function LCSF, a human eye vision absolute contrast sensitivity function ACSF, a feeling brightness contrast function FBCS and a local average gradient sensitivity function LGSF to obtain each grade of the sub-band coefficient of a fused image; and (3) carrying out NSCT inverted conversion on the fused coefficient and reconstructing to obtain the fused image. The multi-modality image fusion method disclosed by the invention meets vision properties of a human eye and improves the quality of the fused image to a great extent; and the multi-modality image fusion method has robustness, is suitable for fusing a plurality of types of image sources of infrared and visible light images, multi-focusing images, remote images and the like, and has a wide application prospect.

The invention discloses a color image color constant method based on a retina vision mechanism, which belongs to the technical field of image processing, and relates to a color image constant technology, in particular to a color image color constant method based on a retina vision mechanism. The invention is based on the current latest optic nerve electrophysiology research achievement to simulatethe retina vision mechanism of human eyes and establish an approaching receptive field single antagonist model to realize the constant color of a deviation color image. The color image color constantmethod based on a retina vision mechanism confirms that the proposed arithmetic is more effective than the current color constant arithmetic through the image validation of more than three hundred different scenes and illumination in the international universal color constant test database; besides, the color image color constant method based on a retina vision mechanism not only can realize theconstant color of the deviation colorful image, but also has the function of reinforcing the image color, thus being more suitable for the visual characteristic of the human eye.

The invention discloses an image enhancing method based on the visual characteristics of human eyes, for example, the human eyes has poor resolution ratio on grayscale when the grayscale of an image is very high or very low, the human eyes has high resolution ratio when the grayscale is moderate. The image enhancing method considers the visual characteristics of the human eyes in the image processing process and improves the defects of the prior image enhancing method. Neighborhood information plays an important role in the image smoothing and denoising process, thus adjacent pixel points canbe used for enhancing processing efficiency in the image processing process. The image enhancing method can smooth and denoise the image and enables the enhanced image to more accord with the visual characteristics of the human eyes.

The invention provides a tongue image environment adaptive color reproduction method of traditional Chinese medicine and belongs to the field of medical image processing. The method comprises the steps of at first, shooting a special color plate image under the same image collection condition and measuring standard colorimetric values of color lumps in the special color plate by colorimeter; establishing a collection process color characterization model through color plate image data in a computer to realize conversion between equipment color space and standard chroma space; and then calculating displayed image chroma tristimulus values by a visual adaption model according to measured obtained image collection environment and displayed environmental parameter which is determined through combination with a background adaption parameter determined by tongue image structure prior knowledge. The method is similar to collection process characterization in that through display of a group of color lump on a displayer and measurement of standard chroma value of displayed color lump, establishment of the collection process color characterization model, and calculation of the display drive value corresponding to chroma tristimulus value. With the tongue image environment adaptive color reproduction method of traditional Chinese medicine , adaptive true color reproduction of tongue images is finally achieved.

The invention aims to provide an improved non-classical receptive field-based bionic contour detection method. The method comprises the following steps of A, inputting a to-be-detected image subjectedto grayscale processing; B, presetting a Gabor filtering function containing multiple direction parameters, performing classical receptive field stimulus response on each pixel point in the to-be-detected image, wherein a corresponding direction serves as an optimal direction of the pixel point; C, building an inhibitory kernel function by utilizing a log function, building a distance weight function through the inhibitory kernel function, and performing convolution on the classical receptive field stimulus response of the pixel point and the distance weight function of the pixel point to obtain inhibitory response of the pixel point; and D, subtracting a product of the inhibitory response of the pixel point and preset inhibitory intensity from the classical receptive field stimulus response of each pixel point, and performing calculation to obtain a final contour value of the pixel point. The detection method overcomes the defects of the prior art and has the characteristics of meeting spatial characteristics of visual receptive field and having a better detection effect.

The invention relates to a multi-band image synchronous fusion and enhancement method. Multi-band image fusion and image super-resolution are synchronously realized; the invention especially relates to a multi-band image synchronous fusion and enhancement method based on improved WGAN-GP. The method comprises the following steps: designing and constructing a generative adversarial network, whereinthe network structure is composed of a conceptual feature extraction (coding) network, a feature combination network and a decoding network, a generative model is obtained through dynamic balance training of a generator and a discriminator, and a fusion enhancement result is obtained. According to the invention, the end-to-end synchronous fusion and enhanced neural network of multi-band images isrealized, so that a high-quality fusion result is fused from a low-resolution source image.

The invention relates to a fusion method for multi-band images, in particular to a self-supervised learning fusion method for multi-band images based on a multi-discriminator generative adversarial network, which comprises the following steps of: designing and constructing a generative adversarial network which consists of a generator and a plurality of discriminators, and a label image is a multi-band source image; wherein a generator network structure is composed of a feature enhancement module and a feature fusion module which are conceived, acquiring a generation model through dynamic balance training of a generator and a discriminator to acquire a multi-band image fusion result. A multi-band image end-to-end self-supervision fusion neural network is realized, a result has better definition and information amount, the detail information is richer, and the method better conforms to the visual characteristics of human eyes.