quickstart edits

Author: HeidiSteen

articles/search/search-get-started-portal-image-search.md

@@ -1,7 +1,7 @@
 ---
 title: Quickstart image search
 titleSuffix: Azure AI Search
-description: Index and query images on Azure AI Search using the Azure portal. Run the Import and vectorize data wizard to vectorize images. Use Search Explorer to query images.
+description: Search for images on Azure AI Search index using the Azure portal. Run the Import and vectorize data wizard to vectorize images, then use Search Explorer to provide an image as your query input.
 
 author: HeidiSteen
 ms.author: heidist
@@ -24,7 +24,7 @@ You need three Azure resources and some sample image files to complete this walk
 > [!div class="checklist"]
 > + Azure Storage to store image files as blobs
 > + Azure AI services multiservice account, used for image vectorization and Optical Character Recognition (OCR)
-> + Azure AI Search for indexing and queries
+> + Azure AI Search for image vectorization during indexing and queries
 
 Sample data consists of image files in the [azure-search-sample-data](https://github.com/Azure-Samples/azure-search-sample-data/tree/main/unsplash-images) repo, but you can use different images and still follow this walkthrough.
 
@@ -100,13 +100,14 @@ Azure AI Vision provides text embeddings, so we'll use that resource for text ve
 
 Use Azure AI Vision to generate a vector representation of the image files. 
 
-In this step, you can also apply AI to extract text from images. The wizard uses OCR from Azure AI services to recognize text in image files. Two more outputs appear in the index when OCR is added to the workflow:
+In this step, you can also apply AI to extract text from images. The wizard uses OCR from Azure AI services to recognize text in image files. 
 
-+ First, the "chunk" field is populated with the OCR-generated string. 
+Two more outputs appear in the index when OCR is added to the workflow:
 
-+ Second, the "text_vector" field is populated with an embedding that represents the string. 
++ First, the "chunk" field is populated with an OCR-generated string of any text in the image. 
++ Second, the "text_vector" field is populated with an embedding that represents the "chunk" string. 
 
-The inclusion of plain text in an index is useful if you want to use relevance features that operate on strings, such as semantic ranking and scoring profiles.
+The inclusion of plain text in the "chunk" field is useful if you want to use relevance features that operate on strings, such as [semantic ranking](semantic-search-overview.md) and [scoring profiles](index-add-scoring-profiles.md).
 
 1. On the **Vectorize your images** page, select the **Vectorize images** checkbox, and then select **AI Vision vectorization**.
 
@@ -143,20 +144,16 @@ The inclusion of plain text in an index is useful if you want to use relevance f
    + Skillset with the following five skills:
 
      + [OCR skill](cognitive-search-skill-ocr.md) recognizes text in image files.
-
      + [Text Merger skill](cognitive-search-skill-textmerger.md) unifies the various outputs of OCR processing.
-
      + [Text Split skill](cognitive-search-skill-textsplit.md) adds data chunking. This skill is built into the wizard workflow.
-
      + [Azure AI Vision multimodal](cognitive-search-skill-vision-vectorize.md) is used to vectorize text generated from OCR.
-
      + [Azure AI Vision multimodal](cognitive-search-skill-vision-vectorize.md) is called again to vectorize images.
 
    + Indexer with field mappings and output field mappings.
 
 ## Check results
 
-Search explorer accepts text, vectors, and images as query inputs. You can drag or select an image into the search area, and it will be vectorized for search. Image vectorization assumes that your index has a vectorizer definition, which **Import and vectorize data** creates from your inputs.
+Search explorer accepts text, vectors, and images as query inputs. You can drag or select an image into the search area, and it will be vectorized for search. Search Explorer vectorizes your image and sends the vector as a query input to the search engine. Image vectorization assumes that your index has a vectorizer definition, which **Import and vectorize data** creates from your inputs.
 
 1. In the Azure portal, under **Search Management** and **Indexes**, select the index your created. An embedded Search Explorer is the first tab.
 
@@ -170,7 +167,7 @@ Search explorer accepts text, vectors, and images as query inputs. You can drag
 
    :::image type="content" source="media/search-get-started-portal-images/image-search.png" alt-text="Screenshot of search results.":::
 
-   The top match should be the image you searched for. Because a vector search matches on similar vectors, the search engine returns any document that is sufficiently similar to the query input. You can switch to JSON view for more advanced queries that include relevance tuning.
+   The top match should be the image you searched for. Because a [vector search](vector-search-overview.md) matches on similar vectors, the search engine returns any document that is sufficiently similar to the query input, up to *k*-number of results. You can switch to JSON view for more advanced queries that include relevance tuning.
 
 1. Try other query options to compare search outcomes:
 

articles/search/whats-new.md

@@ -23,7 +23,7 @@ ms.custom:
 
 | Item                         | Type |  Description |
 |-----------------------------|------|--------------|
-| [Image search in the Azure portal](search-get-started-portal-image-search.md) | Feature | Search explorer now supports image search. In a vector index that has vectorized image content, you can drop images into Search Explorer to query for matching image content.
+| [Image search in the Azure portal](search-get-started-portal-image-search.md) | Feature | Search explorer now supports image search. In a vector index that has vectorized image content, you can drop images into Search Explorer to query for a match.
 
 ## May 2024