This workflow follows the Editimage → HTTP Request recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"nodes": [
{
"id": "6359f725-1ede-4b05-bc19-05a7e85c0865",
"name": "When clicking \"Test workflow\"",
"type": "n8n-nodes-base.manualTrigger",
"position": [
680,
292
],
"parameters": {},
"typeVersion": 1
},
{
"id": "9e1e61c7-f5fd-4e8a-99a6-ccc5a24f5528",
"name": "Fetch Source Image",
"type": "n8n-nodes-base.httpRequest",
"position": [
1000,
292
],
"parameters": {
"url": "={{ $json.source_image }}",
"options": {}
},
"typeVersion": 4.2
},
{
"id": "9b1b94cf-3a7d-4c43-ab6c-8df9824b5667",
"name": "Split Out Results Only",
"type": "n8n-nodes-base.splitOut",
"position": [
1428,
323
],
"parameters": {
"options": {},
"fieldToSplitOut": "result"
},
"typeVersion": 1
},
{
"id": "fcbaf6c3-2aee-4ea1-9c5e-2833dd7a9f50",
"name": "Filter Score >= 0.9",
"type": "n8n-nodes-base.filter",
"position": [
1608,
323
],
"parameters": {
"options": {},
"conditions": {
"options": {
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "367d83ef-8ecf-41fe-858c-9bfd78b0ae9f",
"operator": {
"type": "number",
"operation": "gte"
},
"leftValue": "={{ $json.score }}",
"rightValue": 0.9
}
]
}
},
"typeVersion": 2
},
{
"id": "954ce7b0-ef82-4203-8706-17cfa5e5e3ff",
"name": "Crop Object From Image",
"type": "n8n-nodes-base.editImage",
"position": [
2080,
432
],
"parameters": {
"width": "={{ $json.box.xmax - $json.box.xmin }}",
"height": "={{ $json.box.ymax - $json.box.ymin }}",
"options": {
"format": "jpeg",
"fileName": "={{ $binary.data.fileName.split('.')[0].urlEncode()+'-'+$json.label.urlEncode() + '-' + $itemIndex }}.jpg"
},
"operation": "crop",
"positionX": "={{ $json.box.xmin }}",
"positionY": "={{ $json.box.ymin }}"
},
"typeVersion": 1
},
{
"id": "40027456-4bf9-4eea-8d71-aa28e69b29e5",
"name": "Set Variables",
"type": "n8n-nodes-base.set",
"position": [
840,
292
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "9e95d951-8530-4a80-bd00-6bb55623a71f",
"name": "CLOUDFLARE_ACCOUNT_ID",
"type": "string",
"value": ""
},
{
"id": "66807a90-63a1-4d4e-886e-e8abf3019a34",
"name": "model",
"type": "string",
"value": "@cf/facebook/detr-resnet-50"
},
{
"id": "a13ccde6-e6e3-46f4-afa3-2134af7bc765",
"name": "source_image",
"type": "string",
"value": "https://images.pexels.com/photos/2293367/pexels-photo-2293367.jpeg?auto=compress&cs=tinysrgb&w=600"
},
{
"id": "0734fc55-b414-47f7-8b3e-5c880243f3ed",
"name": "elasticsearch_index",
"type": "string",
"value": "n8n-image-search"
}
]
}
},
"typeVersion": 3.3
},
{
"id": "c3d8c5e3-546e-472c-9e6e-091cf5cee3c3",
"name": "Use Detr-Resnet-50 Object Classification",
"type": "n8n-nodes-base.httpRequest",
"position": [
1248,
324
],
"parameters": {
"url": "=https://api.cloudflare.com/client/v4/accounts/{{ $('Set Variables').item.json.CLOUDFLARE_ACCOUNT_ID }}/ai/run/{{ $('Set Variables').item.json.model }}",
"method": "POST",
"options": {},
"sendBody": true,
"contentType": "binaryData",
"authentication": "predefinedCredentialType",
"inputDataFieldName": "data",
"nodeCredentialType": "cloudflareApi"
},
"credentials": {
"cloudflareApi": {
"name": "<your credential>"
}
},
"typeVersion": 4.2
},
{
"id": "3c7aa2fc-9ca1-41ba-a10d-aa5930d45f18",
"name": "Upload to Cloudinary",
"type": "n8n-nodes-base.httpRequest",
"position": [
2380,
380
],
"parameters": {
"url": "https://api.cloudinary.com/v1_1/daglih2g8/image/upload",
"method": "POST",
"options": {},
"sendBody": true,
"sendQuery": true,
"contentType": "multipart-form-data",
"authentication": "genericCredentialType",
"bodyParameters": {
"parameters": [
{
"name": "file",
"parameterType": "formBinaryData",
"inputDataFieldName": "data"
}
]
},
"genericAuthType": "httpQueryAuth",
"queryParameters": {
"parameters": [
{
"name": "upload_preset",
"value": "n8n-workflows-preset"
}
]
}
},
"credentials": {
"httpQueryAuth": {
"name": "<your credential>"
}
},
"typeVersion": 4.2
},
{
"id": "3c4e1f04-a0ba-4cce-b82a-aa3eadc4e7e1",
"name": "Create Docs In Elasticsearch",
"type": "n8n-nodes-base.elasticsearch",
"position": [
2580,
380
],
"parameters": {
"indexId": "={{ $('Set Variables').item.json.elasticsearch_index }}",
"options": {},
"fieldsUi": {
"fieldValues": [
{
"fieldId": "image_url",
"fieldValue": "={{ $json.secure_url.replace('upload','upload/f_auto,q_auto') }}"
},
{
"fieldId": "source_image_url",
"fieldValue": "={{ $('Set Variables').item.json.source_image }}"
},
{
"fieldId": "label",
"fieldValue": "={{ $('Crop Object From Image').item.json.label }}"
},
{
"fieldId": "metadata",
"fieldValue": "={{ JSON.stringify(Object.assign($('Crop Object From Image').item.json, { filename: $json.original_filename })) }}"
}
]
},
"operation": "create",
"additionalFields": {}
},
"credentials": {
"elasticsearchApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "292c9821-c123-44fa-9ba1-c37bf84079bc",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
620,
120
],
"parameters": {
"color": 7,
"width": 541.1455500767354,
"height": 381.6388867600897,
"content": "## 1. Get Source Image\n[Read more about setting variables for your workflow](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.set)\n\nFor this demo, we'll manually define an image to process. In production however, this image can come from a variety of sources such as drives, webhooks and more."
},
"typeVersion": 1
},
{
"id": "863271dc-fb9d-4211-972d-6b57336073b4",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
1180,
80
],
"parameters": {
"color": 7,
"width": 579.7748008857744,
"height": 437.4680103498263,
"content": "## 2. Use Detr-Resnet-50 Object Classification\n[Learn more about Cloudflare Workers AI](https://developers.cloudflare.com/workers-ai/)\n\nNot all AI workflows need an LLM! As in this example, we're using a non-LLM vision model to parse the source image and return what objects are contained within. The image search feature we're building will be based on the objects in the image making for a much more granular search via object association.\n\nWe'll use the Cloudflare Workers AI service which conveniently provides this model via API use."
},
"typeVersion": 1
},
{
"id": "b73b45da-0436-4099-b538-c6b3b84822f2",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
1800,
260
],
"parameters": {
"color": 7,
"width": 466.35460775498495,
"height": 371.9272151757119,
"content": "## 3. Crop Objects Out of Source Image\n[Read more about Editing Images in n8n](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.editimage)\n\nWith our objects identified by their bounding boxes, we can \"cut\" them out of the source image as separate images."
},
"typeVersion": 1
},
{
"id": "465bd842-8a35-49d8-a9ff-c30d164620db",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
2300,
180
],
"parameters": {
"color": 7,
"width": 478.20345439832454,
"height": 386.06196032653685,
"content": "## 4. Index Object Images In ElasticSearch\n[Read more about using ElasticSearch](https://docs.n8n.io/integrations/builtin/app-nodes/n8n-nodes-base.elasticsearch)\n\nBy storing the newly created object images externally and indexing them in Elasticsearch, we now have a foundation for our Image Search service which queries by object association."
},
"typeVersion": 1
},
{
"id": "6a04b4b5-7830-410d-9b5b-79acb0b1c78b",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
1800,
-220
],
"parameters": {
"color": 7,
"width": 328.419768654291,
"height": 462.65463700396174,
"content": "Fig 1. Result of Classification\n"
},
"typeVersion": 1
},
{
"id": "8f607951-ba41-4362-8323-e8b4b96ad122",
"name": "Fetch Source Image Again",
"type": "n8n-nodes-base.httpRequest",
"position": [
1880,
432
],
"parameters": {
"url": "={{ $('Set Variables').item.json.source_image }}",
"options": {}
},
"typeVersion": 4.2
},
{
"id": "6933f67d-276b-4908-8602-654aa352a68b",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
220,
120
],
"parameters": {
"width": 359.6648027457353,
"height": 352.41026669883723,
"content": "## Try It Out!\n### This workflow does the following:\n* Downloads an image\n* Uses an object classification AI model to identify objects in the image.\n* Crops the objects out from the original image into new image files.\n* Indexes the image's object in an Elasticsearch Database to enable image search.\n\n### Need Help?\nJoin the [Discord](https://discord.com/invite/XPKeKXeB7d) or ask in the [Forum](https://community.n8n.io/)!\n\nHappy Hacking!"
},
"typeVersion": 1
},
{
"id": "35615ed5-43e8-43f0-95fe-1f95a1177d69",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
800,
280
],
"parameters": {
"width": 172.9365918827757,
"height": 291.6881468483679,
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ud83d\udea8**Required**\n* Set your variables here first!"
},
"typeVersion": 1
}
],
"connections": {
"Set Variables": {
"main": [
[
{
"node": "Fetch Source Image",
"type": "main",
"index": 0
}
]
]
},
"Fetch Source Image": {
"main": [
[
{
"node": "Use Detr-Resnet-50 Object Classification",
"type": "main",
"index": 0
}
]
]
},
"Filter Score >= 0.9": {
"main": [
[
{
"node": "Fetch Source Image Again",
"type": "main",
"index": 0
}
]
]
},
"Upload to Cloudinary": {
"main": [
[
{
"node": "Create Docs In Elasticsearch",
"type": "main",
"index": 0
}
]
]
},
"Crop Object From Image": {
"main": [
[
{
"node": "Upload to Cloudinary",
"type": "main",
"index": 0
}
]
]
},
"Split Out Results Only": {
"main": [
[
{
"node": "Filter Score >= 0.9",
"type": "main",
"index": 0
}
]
]
},
"Fetch Source Image Again": {
"main": [
[
{
"node": "Crop Object From Image",
"type": "main",
"index": 0
}
]
]
},
"When clicking \"Test workflow\"": {
"main": [
[
{
"node": "Set Variables",
"type": "main",
"index": 0
}
]
]
},
"Use Detr-Resnet-50 Object Classification": {
"main": [
[
{
"node": "Split Out Results Only",
"type": "main",
"index": 0
}
]
]
}
}
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
cloudflareApielasticsearchApihttpQueryAuth
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
Users gain a custom image search engine that automatically detects and crops specific objects from images, enabling efficient organisation and retrieval of visual content without manual tagging. This workflow suits content creators, e-commerce managers, or developers handling large image libraries who need precise, AI-powered detection to streamline workflows. The key step involves using object classification via a DETR-ResNet-50 model to identify and isolate elements, followed by uploading cropped results to Cloudinary for storage and Elasticsearch for searchable indexing.
Employ this workflow when building an event-driven system to process incoming images in real-time, such as analysing user uploads for product catalogues or media archives. Avoid it for simple keyword-based searches or when dealing with low-confidence detections below 0.9 score, as it prioritises high-accuracy crops. Common variations include integrating additional filters for specific object types or extending Elasticsearch queries for advanced faceted searches.
About this workflow
Build Your Own Image Search Using Ai Object Detection, Cdn And Elasticsearchbuild Your Own Image Search Using Ai Object Detection, Cdn And Elasticsearch. Uses manualTrigger, httpRequest, splitOut, editImage. Event-driven trigger; 17 nodes.
Source: https://github.com/Zie619/n8n-workflows — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This template can be used to generate research ideas from PDF scientific papers based on the content gaps found in text using the InfraNodus knowledge graph GraphRAG knowledge graph representation.
Summarize YouTube Videos with Google Gemini 2.5. Uses formTrigger, httpRequest. Event-driven trigger; 10 nodes.
This workflow takes two inputs, YouTube video URL (required) and a description of what information to extract from the video. If the description/"what you want" field is left empty, the default prompt
IntelliX.AI - Editorial Automatizado v2. Uses postgres, rssFeedRead, openAi, httpRequest. Event-driven trigger; 62 nodes.
Clone_Viral_TikToks_with_AI_Avatars___Auto_Post_to_9_Platforms_using_Perplexity___Blotato. Uses httpRequest, telegramTrigger, openAi, googleSheets. Event-driven trigger; 42 nodes.