Tausif (2)
- 1. COLOR AND TEXTURE FEATURES OF IMAGE INDEXING AND RETRIEVAL
- 2. INTRODUCTION Application of world wide web and the internet is increasing exponentially, and with it the amount of digital image data accessible to the users. A huge amount of image databases are added every minute and so is the need effective and efficient image retrieval systems. There are many features of content based image retrieval but mainly four of them are the main features. They are color, texture, shape and spatial properties. Spatial properties, however, are implicitly taken into account so the main features to investigate are color , texture and shape.
- 3. COLOR HISTOGRAM In content based image retrieval, color descriptor has been one of the first choices because if one chooses a proper representation , it can be partially reliable even in the presence of changes in lighting, view angle and scale. In image retrieval, the color histogram is the most commonly used global color feature. It denotes the probability of the intensities of the three color channels. Typical characterization of color composition is done by color histogram. In the year 1991 a method was proposed, called color indexing, which identifies the object using color histogram indexing.. The color histogram is obtained by counting the number of times each color occurs in the image array. Histogram is invariant to translation and rotation of the image plane, and change only slowly under change of angle of view.
- 4. NEED OF TRANSFORM Mathematical transformation are applied to signals to obtain a further information from that signal that is not really available in the raw signal. There are number of transformation that can be applied, among which the Fourier transform are probably most important. Most of the signal are TIME-DOMAIN in there raw format that is whatever the signal is measuring is a function of time. A time- amplitude representation of the signal is obtained when time-domain signal is plotted. In many cases , the most distinguished information is hidden in the frequency content of the signal. Intuitively, frequency is something to do with the rate of change of a mathematical or physical variable. The frequency content of the signal is measured by the help of Fourier transform. If the FT of the signal in time domain is taken the frequency-amplitude representation of that signal is obtained.
- 5. WAVELET TRANSFORM The wavelet transform is the transform of the type which provides time-frequency representation. Often times a particular spectral components occurring at any instant can be of particular interest. In these cases it may be very beneficial to know the time intervals these particular spectral components occur. Wavelet transform is capable of providing the time and frequency information, hence giving a time – frequency representation of the signal. The wavelet transform is being developed to overcome resolution related problems.
- 6. HOW THIS WORKS Suppose we have a signal which has frequencies up-t0 1000Hz. In the first stage the signal is splited into two parts by passing the signal from a highpass and a lowpass filter. Filter should satisfy some certain conditions known as admissibility condition resulting in two different version of the same signal: portion of the signal corresponding to 0-500 Hz(low pass portion), and 500-1000Hz (high pass portion). Usually low pass portion is used and the operation is called decomposition. Considering low pass portion 3sets of data, each corresponding to the same signal at frequencies 0-250Hz, 250-500Hz,500-1000Hz.
- 7. Then again we take the lowpass portion and pass it through low and high pass filters; now there are 4 sets of signals corresponding to 0-125Hz,125-250Hz,250-500Hz, and 500- 1000Hz. This process is continued until the signal is decomposed to a pre-defined certain level. By this we have a bunch of signals, which actually represent the same signal, but all corresponding to different frequency bands. Each signal corresponds to a particular frequency band and are plotted together on a 3-D graph in which time will be in one axis , frequency in the second and amplitude in the third. This will show which frequency will exist at which time.
- 8. MULTIRESOLUTION ANALYSIS It is possible to analyze any signal by using an alternative approach called the “multiresolution analysis” . MRA is designed to give good time resolution and poor resolution at high frequency and good frequency resolution and poor time resolution at low frequencies. This approach makes sense especially when the signal has high frequency components for short durations and low frequency components for long durations
- 9. INTRODUCTION TO CBIR Now a days ,CBIR (content based image retrieval) is a hotspot in “Digital image processing techniques”. There is a growing interest in CBIR because of the limitations inherent in metadata-based systems, as well as the large range of possible uses for efficient image retrieval. The term ‘content’ in this context might refers to colors, shapes, textures, or any other information that can be derived from the image itself.
- 10. APPROACHES TO IMAGE RETRIEVAL There are two approaches to image retrieval: Text-Based Approach Content-Based Approach Text –Based approach has some obvious shortcomings as each person can have different perception for each textual description. It is also time consuming when dealing with very large databases. Content based retrieval of visual data requires a paradigm that differs significantly from both traditional databases and text based image understanding systems.