Basic Concept
Image enhancement is highlighting some of the information in an image at a specific requirement, while weakening some information processing methods. For example, the original image (image) is copied into a positive particle and a negative film, and the diaphragm of the two different properties is precisely overlap, two diaphragms The distance between mutual misplaced, so that a corresponding image has a slightly misplaced striped image, most of its images are offset, and a bright line (or dark line) appears in the edge portion, reaching outgoing image from the background The display effect of the boundary line makes the image enhancements. Image enhanced by image enhancement is very effective for certain specific applications. Common image enhancements have histogram modifications, smoothing, image sharpening, color processing technology, etc. Image enhancements include contrast enhancement of images, brightness enhancement and contour increase.
Edge and Edge Detection
image has edges. Although the computer is very large, it is impossible to handle images without edges. The edge of the image is the most essential feature of the image. It refers to a collection of pixels that have righteous variations or roof-like changes. Poggio said: "The edge may have no corresponding image of the object (boundary) in the image, but the edge has a very satisfactory nature, which greatly reduces the information to be processed but retained The shape information of the object in the image. " The edges are widely present between the objects and the background, and the objects are between the objects. Therefore, it is an important feature of image segmentation. The edge detection of the image (or the edge enhancement) is the contour of the image to make the image of the image. Image processing method. It is an important area processing technology. The edge detection is generally required before the image is extracted, and then binaryization is performed. Edge detection will protrude the edge of the image, the image area other than the edge is usually weakened or even completely removed. The brightness of the processed boundary is proportional to the brightness of the edge around the edge of the original.
Edge Enhance Imaging and Progress
During general imaging, we have obtained the image of the field of light intensity information. In many applications, imaging focuses on the characteristics details of the field, such as edge profile, etc., which requires edge enhancement processing on the field to highlight the desired edge profile. In order to obtain an edge enhanced image of the object field, one means is to image processing through the computer, but this method is limited to the information quantity of the obtained intensity image itself and the resolution and quality of the image. Factors, not available information; another method is to directly obtain edge enhancement information after processing information on the field of imaging system, which requires different optical components to apply different optical components in imaging systems. Processing, enhancement of detail information such as amplitude and phase of the field can be realized. In fact, these two methods are the later processing one is the pre-processing, and the theoretical basis applied is essentially consistent. In the spatial domain, the edge portion of the field is generally referred to as the boundary or reconstitution of the object field, and the region of the inner region is, and the change area of the object field. Generally, the object field can be described by a function, and if the function is different, the differential value of the variable region of the field field is small, and the differential value of the field edge mutant region is large. From this, it can be seen that the edge enhancement information of the field can be obtained by differential processing on the field. At the spectrum domain, the edge portion of the field is referred to, refers to the high frequency part of the field spatial spectrum, not the edge region, the low frequency portion of the spatial spectrum. It can be seen that by regulating or selecting the spectrum of the field, the edge enhancement information of the field can be obtained. In recent years, the image processing and acquisition is widely studied by obtaining image edge enhancement in the spatial domain or spectral domain operation, has been widely used in image processing, astronomical detection, pattern identification, etc.. Among them, many of them are utilizing spatial filtering to achieve transform or spectrum modulation on the object field to obtain edge enhancements, for example, with a filter that is transformed with the coordinate, composite sinusoidal grating, etc., the composite sinusoidal grating, etc. can be achieved to obtain a differential field to obtain an edge Enhanced; the low frequency component of the field can be removed or suppressed by the high-pass filter (due to most of the energy field is concentrated on its low frequency component, the high-pass filtering will make the energy loss of the field), so that the image of the image Improve, get a clearer edge; you can also perform Hilbert transform on the field, especially the proposal and development of vortex filtering, can hardly lose the energy of the field energy, so that this method has a large Application prospects. In addition to obtaining edge enhancement using spatial filtering, the Laplas differential reconstruction of the field field information recorded under different distances or conditions can also be obtained by using holographic technology.
Image Enhancement Algorithm Overview
With the continuous development of imaging technology and computer technology, digital images have become an indispensable part of people's typical life, various imaging means And equipment is used in digital entertainment, identification, intelligent transportation, medical condition testing, military supervision, etc. Due to the insufficient nightlight, the fog, dust is affected, the burst of emergencies such as cameras in the image, or the target movement in the scene, will cause damage to the captured image or image sequence, and there is imaging blur, noise pollution and It is especially important to have an uneven exposure, for this purpose, image enhancement is particularly important. Comprehensive analysis related literature, image enhancement technology can be independently returned to airspace method and frequency domain method. Directly administer the pixel value to obtain enhanced effects, such as grayscale transformation, histoforming, histoforming method, etc. In the frequency domain method, the image is first converted to the frequency domain, that is, the first step is to perform Fourier Transform. In the frequency enhancement method, all enhanced operators are performed after the Fourier transform is performed, and then the Fourier reverse transform is performed to obtain an enhanced image. These enhanced operators are commonly used to adjust image brightness, contrast, or grayscale distribution, and the pixel values of the output image as the result can be changed according to the applied transformation function. The airspace method has a simple theoretical, and the algorithm has a low complexity, which can be widely used in real-time field image enhancement, and the disadvantage is lack of strong adaptive ability. The frequency domain image enhancement method relies on the frequency domain information conversion processing, which acts on the corresponding transform coefficient. Image enhancement methods of these frequency domains have the advantages of low complexity, corresponding transformation and characteristics. It is the main disadvantage that it is not possible to satisfy all parts of the image, and the algorithm is difficult to deal with adaptive Image enhancement.
With people's increasing demand for high quality images, some details of the scenes in the image color and images are getting higher and higher, how to get high quality images and how to carry low quality degraded images Enhance to achieve better application goals, increasingly become the focus of researchers, and bringing more visual perceived color images and high dynamic range images with higher image detail quality to people more Research interest.
Implementation
For example using related mask technology, copy the original image (image) into a positive film and a negative film, and make two different properties The diaphragm is precisely overlapping. When the exposure is printed, the two diaphragms are mutated with a small distance, so that a corresponding image has a slight misplaced "striped" image, and most of its images are unable to cancel, and A bright line (or dark line) appears in the edge portion, reaching the display effect of protruding the image boundary line from the background to enhance the image. Edge enhancements can also be implemented by other methods or computer processing.