Aus, and thread, I have the ideal anxiety displacement exercise for a day like today:
I am a fisheries scientist - I count fish. I want eventually to use digital cameras to identify and measure individual fish as they come aboard vessels, and go down a shute.
In order to design a sampling scheme, I first have to get a lot of photos of individual fish as they slide down this shute. The photos don't need to be consecutive and it doesn't matter if a large proportion of them cannot be used, as long as I can get the species and size of, say every tenth fish or so. I am talking about large numbers, typically 20 metric tons with an average fish of around 4kg, so 5 000 fish, 500 pictures. These are processed in 30- 45 minutes. There is a break of at least several hours to recharge batteries, exchange cards and possibly transfer images.
The resolution does not have to be magazine quality. I intend to experiment to determine the minimum resolution necessary for identification. We need to be able to see the shape of the pectoral fins and to see some bands of alternating dark and light stripes. The fish are tuna, for those who care.
I intend to buy a few cameras to get those pictures. They will enable me to design a sampling scheme that, hopefully, will require fewer shots and lower resolution. Once I've done that I'll probably come back asking for advice concerning the most robust camera with the required capability. The idea will be to put one on every boat. WE have an observer on every boat, who would change cards, etc after each fishing operation. We hope to be able to transmit this information via satellite to improve the timeliness of our data. Boats can stay at sea for three months.
There's a lot of expertise and experience on this thread. I own some Sandisk, but no digital camera. I'd be grateful for any suggestions. Dare I ask here what people think of the SONY digital-video? I assume this uses the memory-stick.
Ashley |
