{
  "id": 11459,
  "title": "Striking similarity in images",
  "url": "/competitions/datasciencebowl/discussion/11459",
  "author_name": "",
  "post_date": "2015-01-09T04:00:05.090Z",
  "votes": 9,
  "comment_count": 4,
  "views": 2130,
  "content": "<p>Hi,</p>\n<p>There is so much similarity between images of different classes like shrimp caridean and zoea. I am curious to know if marine experts did actually distinguish between them based on same images that are provided for this competition? If yes I would like some insight on how did they distinguish between the attached images.</p>\n<p>Thanks</p>",
  "messages": [
    {
      "id": "62479",
      "postDate": "01/09/2015 04:00:05",
      "content": "<p>Hi,</p>\n<p>There is so much similarity between images of different classes like shrimp caridean and zoea. I am curious to know if marine experts did actually distinguish between them based on same images that are provided for this competition? If yes I would like some insight on how did they distinguish between the attached images.</p>\n<p>Thanks</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "62551",
      "postDate": "01/10/2015 03:59:18",
      "content": "<p>In the Plankton_Classification.pdf guide they are both classified as shrimp-like decapods. They sure look identical to me. Maybe that is why they are connected with a&nbsp;dotted line in the family tree.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "62552",
      "postDate": "01/10/2015 04:34:32",
      "content": "<p>True,</p>\n<p>But we are required to provide the probability of both classes separately for any test image. &nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "62584",
      "postDate": "01/10/2015 20:37:17",
      "content": "<p>Hi @aseem,</p>\n<p>The term &quot;zoea&quot; refers to the early life history stage of crustaceans - including shrimp. Here we did our best to distinguish within a developmental sequence -- the early life stages of shrimps were put into the zoea category. Because of the nature of separating a continual sequence into discrete groups, there are going to be images on the boundary where the distinction between groups are less clear.&nbsp;</p>\n<p>For the caridean shrimp vs zoea groups, there are similarities between the two -- and we acknowledged these similarities in the plankton classification chart.&nbsp;</p>\n<p>Also - refer to the FAQs in the &quot;get the data&quot; page where I talk about the intrinsic uncertainty in classifying ambiguous images and the reliability of our classifications.</p>\n\n<p>cheers,</p>\n<p>Jessica</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "64502",
      "postDate": "02/18/2015 15:37:13",
      "content": "<p>If they look identical, then we should assign equal probability to them.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 62551,
      "author_name": "jeffhebert",
      "author_url": "",
      "post_date": "01/10/2015 03:59:18",
      "content": "<p>In the Plankton_Classification.pdf guide they are both classified as shrimp-like decapods. They sure look identical to me. Maybe that is why they are connected with a&nbsp;dotted line in the family tree.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 62552,
      "author_name": "aseem1",
      "author_url": "",
      "post_date": "01/10/2015 04:34:32",
      "content": "<p>True,</p>\n<p>But we are required to provide the probability of both classes separately for any test image. &nbsp;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 62584,
      "author_name": "jessicaluo",
      "author_url": "",
      "post_date": "01/10/2015 20:37:17",
      "content": "<p>Hi @aseem,</p>\n<p>The term &quot;zoea&quot; refers to the early life history stage of crustaceans - including shrimp. Here we did our best to distinguish within a developmental sequence -- the early life stages of shrimps were put into the zoea category. Because of the nature of separating a continual sequence into discrete groups, there are going to be images on the boundary where the distinction between groups are less clear.&nbsp;</p>\n<p>For the caridean shrimp vs zoea groups, there are similarities between the two -- and we acknowledged these similarities in the plankton classification chart.&nbsp;</p>\n<p>Also - refer to the FAQs in the &quot;get the data&quot; page where I talk about the intrinsic uncertainty in classifying ambiguous images and the reliability of our classifications.</p>\n\n<p>cheers,</p>\n<p>Jessica</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 64502,
      "author_name": "oneoff",
      "author_url": "",
      "post_date": "02/18/2015 15:37:13",
      "content": "<p>If they look identical, then we should assign equal probability to them.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "62479": "",
    "62551": "",
    "62552": "",
    "62584": "",
    "64502": ""
  },
  "source": "meta"
}