Imge processing

[Altera_Forum]

What do you mean by "Image processing": 

 1. simple image manipulation or 2. real image processing (filtering, thresholding, boudnary detection, segmentation, compression etc)?  I mean, is this like ms-paint or you want it for academic purposes? If answer is 2, the only thing you need to know from Visual C++ is how to load images from files (bmp,jpeg etc). By the time you have the image stored in a buffer (say BUTE*, or short*) what you actually need is knowledge of basic C and very good knowledge of image processing algorithms. I myself did in my final project for the University a medical image processing tool (for image segmentation) using VC++. All algorithms where developed using plain C.

[Altera_Forum]

Quick introduction: 

The first step towards designing an image analysis system is digital image acquisition using sensors in optical or thermal wavelengths. a twodimensional image that is recorded by these sensors is the mapping of the three-dimensional visual world. The captured two dimensional signals are sampled and quantized to yield digital images. Sometimes we receive noisy images that are degraded by some degrading mechanism. One common source of image degradation is the optical lens system in a digital camera that acquires the visual information. If the camera is not appropriately focused then we get blurred images. Here the blurring mechanism is the defocused camera. Very often one may come across images of outdoor scenes that were procured in a foggy environment. Thus any outdoor scene captured on a foggy winter morning could invariably result into a blurred image. In this case the degradation is due to the fog and mist in the atmosphere, and this type of degradation is known as atmospheric degradation. In some other cases there may be a relative motion between the object and the camera. Thus if the camera is given an impulsive displacement during the image capturing interval while the object is static, the resulting image will invariably be blurred and noisy. In some of the above cases, we need appropriate techniques of refining the images so that the resultant images are of better visual quality, free from aberrations and noises. Image enhancement, filtering, and restoration have been some of the important applications of image processing since the early days of the field. Segmentation is the process that subdivides an image into a number of uniformly homogeneous regions. Each homogeneous region is a constituent part or object in the entire scene. In other words, segmentation of an image is defined by a set of regions that are connected and nonoverlapping, so that each pixel in a segment in the image acquires a unique region label that indicates the region it belongs to. Segmentation is one of the most important elements in automated image analysis, mainly because a t this step the objects or other entities of interest are extracted from an image for subsequent processing, such as description and recognition. For example, in case of an aerial image containing the ocean and land, the problem is to segment the image initially into two parts-land segment and water body or ocean segment. Thereafter the objects on the land part of the scene need to be appropriately segmented and subsequently classified. After extracting each segment; the next task is to extract a set of meaningful features such as texture, color, and shape. These are important measurable entities which give measures of various properties of image segments. Some of the texture properties are coarseness, smoothness, regularity, etc., while the common shape descriptors are length, breadth, aspect ratio, area, location, perimeter, compactness, etc. Each segmented region in a scene may be characterized by a set of such features. Finally based on the set of these extracted features, each segmented object is classified to one of a set of meaningful classes. In a digital image of ocean, these classes may be ships or small boats or even naval vessels and a large class of water body. The problems of scene segmentation and object classification are two integrated areas of studies in machine vision. Expert systems, semantic networks, and neural network-based systems have been found to perform such higher-level vision tasks quite efficiently. Another aspect of image processing involves compression and coding of the visual information. With growing demand of various imaging applications, storage requirements of digital imagery are growing explosively. Compact representation of image data and their storage and transmission through communication bandwidth is a crucial and active area of development today. Interestingly enough, image data generally contain a significant amount of superfluous and redundant information in their canonical representation. Image compression techniques helps to reduce the redundancies in raw image data in order to reduce the storage and communication bandwidth.