Tuesday, April 14, 2009
42、炒波菜时不宜加盖 43、炒肉片：肉切成薄片加酱油、黄油、淀粉，打入一个鸡蛋，拌匀，炒散；等肉片变色后，再加佐料稍炒几下，肉片味美、鲜嫩 44、炒牛肉丝：切好，用盐、糖、酒、生粉(或鸡蛋)拌一下，加上生油泡腌，30分钟后再炒，鲜嫩可口
Sunday, April 12, 2009
The next step for me is learning how to write on Latex. It seems harder than Office Word and OpenOffice Writer. Although Latex is much stronger than both of them mentioned above, I trid to face it before, for its complicate grammer and the huge number of identifiers and macros. Today, the comfortable day needs to be terminated...
Friday, April 10, 2009
However, today, I really felt a little tired, both mentally and physically. The system, my first design, called 'hybrid framework for action recognition’, was getting bigger and more complicated. I spent more than 24 hours to finish and debug a function, which has only little weight in the framework and is a very small segment of the system, and I had worked out 25 functions like that... no progress, no result, no graphic output... I feel tired now. An upset staff came into my mind then, IBM OS/2. OS/2 is such a system, designed by a group of talent scientists and engineers, but they spent too much resource on trivial details and failed in competition at last...
I do not want my first design dead. However, I could do nothing more but add more code on it to maintain its life... More codes, more complicated, more errors, more codes to fix it, again and again, dead loop…
The diagnostic performance of a test, or the accuray of a test to discriminate diseased cases from normal cases is evaluated using Receiver Operating Characteristic (ROC) curve analysis (Metz, 1978; Zweig & Campbell, 1993). ROC curves can also be used to compare the diagnostic performance of two or more laboratory or diagnostic tests (Griner et al., 1981).
When you consider the results of a particular test in two populations, one population with a disease, the other population without the disease, you will rarely observe a perfect separation between the two groups. Indeed, the distribution of the test results will overlap, as shown in the following figure.
For every possible cut-off point or criterion value you select to discriminate between the two populations, there will be some cases with the disease correctly classified as positive (TP = True Positive fraction), but some cases with the disease will be classified negative (FN = False Negative fraction). On the other hand, some cases without the disease will be correctly classified as negative (TN = True Negative fraction), but some cases without the disease will be classified as positive (FP = False Positive fraction).
Schematic outcomes of a test
The different fractions (TP, FP, TN, FN) are represented in the following table.
|Positive||True Positive (TP)||a||False Positive (FP)||c||a + c|
|Negative||False Negative (FN)||b||True Negative (TN)||d||b + d|
|Total||a + b||c + d|
The following statistics can be defined:
- Sensitivity: probability that a test result will be positive when the disease is present (true positive rate, expressed as a percentage).
= a / (a+b)
- Specificity: probability that a test result will be negative when the disease is not present (true negative rate, expressed as a percentage).
= d / (c+d)
- Positive likelihood ratio: ratio between the probability of a positive test result given the presence of the disease and the probability of a positive test result given the absence of the disease, i.e.
= True positive rate / False positive rate = Sensitivity / (1-Specificity)
- Negative likelihood ratio: ratio between the probability of a negative test result given the presence of the disease and the probability of a negative test result given the absence of the disease, i.e.
= False negative rate / True negative rate = (1-Sensitivity) / Specificity
- Positive predictive value: probability that the disease is present when the test is positive (expressed as a percentage).
= a / (a+c)
- Negative predictive value: probability that the disease is not present when the test is negative (expressed as a percentage).
= d / (b+d)
Sensitivity and specificity versus criterion value
When you select a higher criterion value, the false positive fraction will decrease with increased specificity but on the other hand the true positive fraction and sensitivity will decrease:
When you select a lower criterion value, then the true positive fraction and sensitivity will increase. On the other hand the false positive fraction will also increase, and therefore the true negative fraction and specificity will decrease.
The ROC curve
In a Receiver Operating Characteristic (ROC) curve the true positive rate (Sensitivity) is plotted in function of the false positive rate (100-Specificity) for different cut-off points. Each point on the ROC plot represents a sensitivity/specificity pair corresponding to a particular decision threshold. A test with perfect discrimination (no overlap in the two distributions) has a ROC plot that passes through the upper left corner (100% sensitivity, 100% specificity). Therefore the closer the ROC plot is to the upper left corner, the higher the overall accuracy of the test (Zweig & Campbell, 1993).
Thursday, April 9, 2009
Ground truth is a term used in cartography, meteorology, analysis of aerial photographs, satellite imagery and a range of other remote sensing techniques in which data are gathered at a distance. Ground truth refers to information that is collected "on location". In remote sensing, this is especially important in order to relate image data to real features and materials on the ground. The collection of ground-truth data enables calibration of remote-sensing data, and aids in the interpretation and analysis of what is being sensed.
More specifically, ground truth may refer to a process in which a pixel on a satellite image is compared to what is there in reality (at the present time) in order to verify the contents of the pixel on the image. In the case of a classified image, it allows supervised classification to help determine the accuracy of the classification performed by the remote sensing software and therefore minimize errors in the classification such as errors of commission and errors of omission.
Ground truth is usually done on site, performing surface observations and measurements of various properties of the features of the ground resolution cells that are being studied on the remotely sensed digital image. It also involves taking geographic coordinates of the ground resolution cell with GPS technology and comparing those with the coordinates of the pixel being studied provided by the remote sensing software to understand and analyze the location errors and how it may affect a particular study.
Ground truth is important in the initial supervised classification of an image. When the identity and location of land cover types are known through a combination of field work, maps, and personal experience these areas are known as training sites. The spectral characteristics of these areas are used to train the remote sensing software using decision rules for classifying the rest of the image. These decision rules such as Maximum Likelihood Classification, Parallelepiped Classification, and Minimum Distance Classification offer different techniques to classify an image. Additional ground truth sites allow the remote sensor to establish an error matrix which validates the accuracy of the classification method used. Different classification methods may have different percentages of error for a given classification project. It is important that the remote sensor chooses a classification method that works best with the number of classifications used while providing the least amount of error.
Ground truth also helps with atmospheric correction. Since images from satellites obviously have to pass through the atmosphere, they can get distorted because of absorption in the atmosphere. So ground truth can help fully identify objects in satellite photos.(PS: All data from Wikipedia)
Sunday, April 5, 2009
In color theory, a tint is the mixture of a color with white, which increases lightness, and a shade is the mixture of a color with black, which reduces lightness. Mixing with any neutral color, including black and white, reduces chroma or colorfulness, while the hue remains unchanged.
When mixing colored light (additive color models), the achromatic mixture of spectrally balanced red, green and blue (RGB) is always white, not gray or black. When we mix colorants, such as the pigments in paint mixtures, a color is produced which is always darker and lower in chroma, or saturation, than the parent colors. This moves the mixed color toward a neutral color―a gray or near-black. Lights are made brighter or dimmer by adjusting their brightness, or energy level; in painting, lightness is adjusted through mixture with white, black or a color's complement.
It is common among some painters to darken a paint color by adding black paint―producing colors called shades―or lighten a color by adding white―producing colors called tints. However it is not always the best way for representational painting, as an unfortunate result is for colors to also shift in hue. For instance, darkening a color by adding black can cause colors such as yellows, reds and oranges, to shift toward the greenish or bluish part of the spectrum. Lightening a color by adding white can cause a shift towards blue when mixed with reds and oranges. Another practice when darkening a color is to use its opposite, or complementary, color (e.g. purplish-red added to yellowish-green) in order to neutralize it without a shift in hue, and darken it if the additive color is darker than the parent color. When lightening a color this hue shift can be corrected with the addition of a small amount of an adjacent color to bring the hue of the mixture back in line with the parent color (e.g. adding a small amount of orange to a mixture of red and white will correct the tendency of this mixture to shift slightly towards the blue end of the spectrum).