III. ANALYSIS OF PROBLEM
The protection of wildlife and forests is a major responsibility of human being. Forest’s officials use to keep track of all movements by each tiger. They used radio collars on tiger shoulder and chips which is in his body to trace the tiger. These both are quite tough jobs. Another method is to track tiger is through their pugmarks. Experience Forest people can identify the tiger by identifies their pugmarks. Forest officials [1] are using radio collars to locate the tigers. These radio collars are heavy (about 1.5-2 kg) and it disturbs the normal behavior of tigers. Collaring the tiger is also a very tough task. Sometimes tigers attack on forest officials during collaring process. This process is very time consuming …show more content…
Methodology
a. Android Camera Mobile: Android mobile phones are used by many person now a days. They provide good features. Our aim is to use the mobile camera to capture the image of tiger pugmark by mobile camera and uploaded it to server for further operations.
b. Image Uploaded To Server – After capturing the image by camera then it is uploaded to server and further operations can be done on server. We have more space on server as compare to the device so the operation and the algorithm will work efficiently on server and not affect on the processing on the mobile phone working.
c. Processing On Server- Server is the space where we can process our pugmark and the data which we are getting can be saved on server and he copy is return on the mobile phone.
d. Data Return By Server On Mobile Phone –
After performing operations on pugmark image the calculated dimensions can be saved on server and a copy for that data has been return on the mobile phone.
e. Compare With Existing Database In A Server –
After calculating dimensions on server the calculated dimensions can be comparing with the existing data on server.
f. If Exist Then Show Result Else New Entry …show more content…
Proposed Algorithm – After doing an literature review on two algorithm[6] SIFT and SURF it is seen that the SIFT has detected more number of features compared to SURF but it is suffered with speed. The SURF is fast and has good performance as the same as SIFT. This is very useful feature of SURF algorithm thus we are implementing this algorithm for image recognition.
The feature finding method is typically composed of 2 steps; first, find the interest points within the image which could contain proposeful structures; this is often typically done by comparing the Difference of Gaussian (DoG) in every location within the image under totally different scales. A significant orientation is additionally calculated when a point is considered a feature point. The second step is to construct the scale invariant descriptor on every interest point found within the previous step. To achieve rotation invariant, we tend to align a rectangle to the main orientation. The dimension of the rectangle is proportional to the size wherever the interest point is detected. The rectangle is then cropped into a 4 by 4 grid. Totally different information’s such as gradient or definite quality of gradient are then subtracted from each of these sub square and composed into the interest point
The protection of wildlife and forests is a major responsibility of human being. Forest’s officials use to keep track of all movements by each tiger. They used radio collars on tiger shoulder and chips which is in his body to trace the tiger. These both are quite tough jobs. Another method is to track tiger is through their pugmarks. Experience Forest people can identify the tiger by identifies their pugmarks. Forest officials [1] are using radio collars to locate the tigers. These radio collars are heavy (about 1.5-2 kg) and it disturbs the normal behavior of tigers. Collaring the tiger is also a very tough task. Sometimes tigers attack on forest officials during collaring process. This process is very time consuming …show more content…
Methodology
a. Android Camera Mobile: Android mobile phones are used by many person now a days. They provide good features. Our aim is to use the mobile camera to capture the image of tiger pugmark by mobile camera and uploaded it to server for further operations.
b. Image Uploaded To Server – After capturing the image by camera then it is uploaded to server and further operations can be done on server. We have more space on server as compare to the device so the operation and the algorithm will work efficiently on server and not affect on the processing on the mobile phone working.
c. Processing On Server- Server is the space where we can process our pugmark and the data which we are getting can be saved on server and he copy is return on the mobile phone.
d. Data Return By Server On Mobile Phone –
After performing operations on pugmark image the calculated dimensions can be saved on server and a copy for that data has been return on the mobile phone.
e. Compare With Existing Database In A Server –
After calculating dimensions on server the calculated dimensions can be comparing with the existing data on server.
f. If Exist Then Show Result Else New Entry …show more content…
Proposed Algorithm – After doing an literature review on two algorithm[6] SIFT and SURF it is seen that the SIFT has detected more number of features compared to SURF but it is suffered with speed. The SURF is fast and has good performance as the same as SIFT. This is very useful feature of SURF algorithm thus we are implementing this algorithm for image recognition.
The feature finding method is typically composed of 2 steps; first, find the interest points within the image which could contain proposeful structures; this is often typically done by comparing the Difference of Gaussian (DoG) in every location within the image under totally different scales. A significant orientation is additionally calculated when a point is considered a feature point. The second step is to construct the scale invariant descriptor on every interest point found within the previous step. To achieve rotation invariant, we tend to align a rectangle to the main orientation. The dimension of the rectangle is proportional to the size wherever the interest point is detected. The rectangle is then cropped into a 4 by 4 grid. Totally different information’s such as gradient or definite quality of gradient are then subtracted from each of these sub square and composed into the interest point