| |
下面引用由li202在 2009/02/15 07:20am 發表的內容：&]Rc!o
©昆蟲論壇 -- 昆蟲論壇 (s
這有兩個問題tS@b#.
1.昆蟲如何判斷空間大小,L;6TM:
2.空間大小是否會影響蟲體大小8.
©昆蟲論壇 -- 昆蟲論壇 j}i
第一個問題'T
圓形容器會不會造成大空間的錯覺??En#w
©昆蟲論壇 -- 昆蟲論壇 n
以容器的形狀作判斷,似乎以人類瞎子摸象的經驗法則來模擬昆蟲,但牠們 ...g9V3'
|
|
W`y
國外研究昆蟲幾何的基本摘要v2:
Preliminary Research and Development of Remote Automatic Insect Image Identification Systemce"7
©昆蟲論壇 -- 昆蟲論壇 >0z
Student: mingyang YuAI-V=4
Supervisor Professor Zuorui Shen%%Y/~u
©昆蟲論壇 -- 昆蟲論壇 PS=7}
Department of Entomology, College of Plant Science and Technology,
China Agricultural University, Beijing 100094, China\r=QP
©昆蟲論壇 -- 昆蟲論壇 t=e<
Human obtain the vast majority of their sensory input through their visual system. Image processing has been put into the use in many applied fileds with huge potential to make complete automation possible, thus overcoming the disadvantages of hand labor. The application of image technology in agriculture, however, is relatively sparse. So the Research and Development of Remote Insect Image Automatic Identification System has a great meaning.k
The study consists of three parts as follows:pJA3Wo
Part 1. Research on automatic acquisition of insect digital images, aiming to capture the digital insect images and video via USB digital camera, using Video for Windows SDK. Video for Windows is a format developed by Microsoft Corporation for storing video and audio information. It was introduced as an SDK separate from a Microsoft OS released in the fall of 1992. VFW became part of the core operating system in Windows 95 and NT3.51. And it will continue to be supported definitely. Files in this format have a .AVI extension. Video for Windows does not require any special hardware, making it the lowest common denominator for multimedia applications. Many multimedia producers use this format, because it allows them to sell their products to the largest base of users.V
Part 2. Research on image processing and analysis, aiming to get the shape features of the insect digital images, through enhancing the digital images, separated the object shape from the background, and detected the image edge.4>J
In this part, we used median filter to enhance the image. Image enhancement techniques are used to increase the signal-to-noise ratio and make certain features easier to see by modifying the colors or intensities of an image. The main image enhancement techniques are intensity adjustment and noise removal. The median filter considers each pixel in the image in turn and looks at its nearby neighbors to decide whether or not it is representative of its surroundings. Instead of simply replacing the pixel value with the mean of neighboring pixel values, the median filter replaces the pixel in the image with the median of those values. The median is calculated by first sorting all the pixel values from the surrounding neighborhood into numerical order and then replacing the pixel being considered with the middle pixel value. (If the neighborhood under consideration contains an even number of pixels, the average of the two middle pixel values is used.).y}TX
The median filter has two main advantages over the mean filter:Jb*%
· The median is a more robust average than the mean and so a very unrepresentative single pixel in a neighborhood will not affect the median value significantly.|9dJ
· Since the median value must actually be the value of one of the pixels in the neighborhood, the median filter does not create new unrealistic pixel values when the filter straddles an edge. For this reason the median filter is much better at preserving sharp edges than the mean filter.|$Y
In general, the median filter allows a great deal of high spatial frequency detail to pass while remaining very effective at removing noise on images where less than half of the pixels in a smoothing neighborhood have been effected./F"f5(
The most powerful edge-detection method in the image processing is mathematical morphology. That image still contains some noise and small size components and needs some further cleaning. Morphological processing is used towards the filtering of small size features and the segmentation of the image. A closing with a structuring element in the horizontal direction followed by an opening with the same structuring element are applied. The results of this operation are shown. A closing followed by an erosion is subtracted from the dilated version of the same image to yield a set of boundaries of the various structures in the image. The objects resulting from the morphological processing were then filtered according to their size, orientation, and elongation.Em+C
The best image segmentation algorithm is minimum fuzziness threshold method based on image entropy.^#I
Based on the segmented image, the boundary of insect image was extracted and geometrical features such as region area, perimeter, eccentricity, shape factor, patch number, Euler's number, roundness, circularity, sphericity and lobation were measured among the training set of the images of the three insect types.#/:Y
Part 3. Research on remote identification, aiming to carry the geometrical features of insect image to server and recognize the insect image by Internet. In this part, we build feature database to manage and save the feature values of insect images. Also, an expert database has been built on the feature database to automatically recognise the insect image. RemoteBug was developed, using a visual programming software, namely, Visual C++ 6.0, under the strategy of OOP (Object Oriented Programming) and provided a user-friendly environment for insect image processing, analysis and recognition.@
Key Words: Insect image, Remote automatic identification, Image processing, Image segmentatoin, Edge detectionns
| |
頂端 
加到"我的最愛"
主題管理：
總固頂 
取消總固頂 
短訊息
查看
搜尋
通訊錄
引用
回覆
給發文者一個讚 
2009/02/15 11:53:19am 
IP: 已設定保密
