Difference between revisions of "ISSS608 2017-18 T3 Assign Yejerla Amrutha Rajeshwari call"
Jump to navigation
Jump to search
(3 intermediate revisions by the same user not shown) | |||
Line 39: | Line 39: | ||
+ | =<font size="5"> Approach </font>= | ||
+ | |||
+ | #Since we have bird recordings submitted by the Ornithology group and also the corresponding quality score, creating a spectograph for individual components of good quality recordings allows for visually profiling each species' calls and songs. | ||
+ | #There are 5 categories in quality score but description of these categories are not provided. Taking few samples in each category and running the autodetec() function on them revealed that category "A" recordings have good quality. | ||
+ | # Atleast 4 recordings for each species- 2 calls, 2 songs(or any other distribution applicable based on the species) are sampled. Components are retrieved and top 5 components based on Signal to Noise ratio are selected. Few most representative spectrographs of each species are documented for comparisons with test recordings. | ||
+ | |||
+ | |||
+ | =<font size="5">Species Call-Song Spectrographs </font>= | ||
+ | |||
{| class="wikitable" style="background-color:#FFFFFF;" width="100%" | {| class="wikitable" style="background-color:#FFFFFF;" width="100%" | ||
Line 201: | Line 210: | ||
|} | |} | ||
+ | |||
+ | =<font size="5"> Species detection for Test files </font>= | ||
{| class="wikitable" style="background-color:#FFFFFF;" width="100%" | {| class="wikitable" style="background-color:#FFFFFF;" width="100%" | ||
Line 222: | Line 233: | ||
||[[File:2-autodetec.ls-th10-env.abs-bp1.10-smo300-midu0.05-mxdu0.5-pw1-p1 - Copy.jpeg|400px]] | ||[[File:2-autodetec.ls-th10-env.abs-bp1.10-smo300-midu0.05-mxdu0.5-pw1-p1 - Copy.jpeg|400px]] | ||
||[[File:Capture_2.PNG|400px]] | ||[[File:Capture_2.PNG|400px]] | ||
− | || | + | || <b><font color="#C00000">Rose crested Blue Pipit along with others</font></b> |
− | |||
|- | |- | ||
Line 287: | Line 297: | ||
||[[File:9-autodetec.ls-th10-env.abs-bp1.10-smo300-midu0.05-mxdu0.5-pw1-p1.jpeg|400px]] | ||[[File:9-autodetec.ls-th10-env.abs-bp1.10-smo300-midu0.05-mxdu0.5-pw1-p1.jpeg|400px]] | ||
||[[File:Capture999.PNG|400px]] | ||[[File:Capture999.PNG|400px]] | ||
− | ||Rose crested Blue Pipit & Ordinary Snape | + | ||<b><font color="#C00000">Rose crested Blue Pipit & Ordinary Snape</font></b> |
Line 351: | Line 361: | ||
|- | |- | ||
|} | |} | ||
+ | |||
+ | '''From the above table, it is evident that at most 2 or 3 recordings belong to Rose crested Blue Pipit. Why did Kaisos officials lie?''' |
Latest revision as of 10:58, 14 July 2018
|
|
|
|
|
|
Approach
- Since we have bird recordings submitted by the Ornithology group and also the corresponding quality score, creating a spectograph for individual components of good quality recordings allows for visually profiling each species' calls and songs.
- There are 5 categories in quality score but description of these categories are not provided. Taking few samples in each category and running the autodetec() function on them revealed that category "A" recordings have good quality.
- Atleast 4 recordings for each species- 2 calls, 2 songs(or any other distribution applicable based on the species) are sampled. Components are retrieved and top 5 components based on Signal to Noise ratio are selected. Few most representative spectrographs of each species are documented for comparisons with test recordings.
Species Call-Song Spectrographs
Species detection for Test files
From the above table, it is evident that at most 2 or 3 recordings belong to Rose crested Blue Pipit. Why did Kaisos officials lie?