I switched the two lines, value is 1, here is the immediate result: edges are a bit not precise since it relies on work path: So I used Select > Grow Like you told me on your first reply + Expand 4px, It's not perfect because it's via paths but it's good enough, i will try on many more pictures and share them Thank you ... View more

Your help is much appreciated! Here is the image. it came from a NEF file type (RAW image) and the dimensions are 6000x4000px I saved it as a jpg because the PSD/TIF file or the NEF file is too big (let me know if you want me still to upload the big file and send you a link to it) But still, save this jpg, Magic Wand on the white part to delete it + add a threshold adjustment layer of 253 then we will be in the same position let me know, thank you! ... View more

the first script: sTT = stringIDToTypeID, dsc = new ActionDescriptor()  dsc.putEnumerated(sTT('colors'), sTT('colors'), sTT('shadows'))  dsc.putInteger(sTT('shadowsUpperLimit'), 0), dsc.putBoolean(sTT('invert'), false)  ref1 = new ActionReference(), ref2 = new ActionReference(), ref1.putClass(sTT('path'))  dsc.putReference(sTT('null'), ref1), ref2.putProperty(sTT('selectionClass'), sTT('selection'))  dsc.putReference(sTT('from'), ref2), dsc.putUnitDouble(sTT('tolerance'), sTT('pixelsUnit'), .5)  executeAction(sTT('colorRange'), dsc), executeAction(sTT('make'), dsc)  len = (sPI = (pI0 = (aD = activeDocument).pathItems[0]).subPathItems).length  for(arr = [], i = 0; i < len; i++) {  for(pts = [], j = 0; j < (pP = sPI.pathPoints).length;) pts.push(pP[j++].anchor);      (sel = aD.selection).select(pts), hst = aD.histogram[0]  if (!arr.length || hst > arr[0][0]) arr[0] = [hst, pts]  pI0.remove(), sel.select(arr[0][1]) returned: ^ it also took a lot of time this script sTT = stringIDToTypeID, dsc = new ActionDescriptor()  dsc.putEnumerated(sTT('colors'), sTT('colors'), sTT('shadows'))  dsc.putInteger(sTT('shadowsUpperLimit'), 0), dsc.putBoolean(sTT('invert'), false)  ref1 = new ActionReference(), ref2 = new ActionReference(), ref1.putClass(sTT('path'))  dsc.putReference(sTT('null'), ref1), ref2.putProperty(sTT('selectionClass'), sTT('selection'))  dsc.putReference(sTT('from'), ref2), dsc.putUnitDouble(sTT('tolerance'), sTT('pixelsUnit'), 1)  executeAction(sTT('colorRange'), dsc), executeAction(sTT('make'), dsc)  len = (sPI = (pI0 = (aD = activeDocument).pathItems[0]).subPathItems).length  for(arr = [], i = 0; i < len; i++) {  for(pts = [], j = 0; j < (pP = sPI.pathPoints).length;) pts.push(pP[j++].anchor);      (sel = aD.selection).select(pts), hst = aD.histogram[0]  if (!arr.length || hst > arr[0][0]) arr[0] = [hst, pts]  pI0.remove(), sel.select(arr[0][1]) returned: this script: sTT = stringIDToTypeID, dsc = new ActionDescriptor()  dsc.putEnumerated(c = sTT('colors'), c, sTT('shadows'))  executeAction(sTT('colorRange'), dsc)    dsc.putInteger(sTT('shadowsUpperLimit'), 0), dsc.putBoolean(sTT('invert'), false)  ref1 = new ActionReference(), ref2 = new ActionReference(), ref1.putClass(sTT('path'))  dsc.putReference(sTT('null'), ref1), ref2.putProperty(sTT('selectionClass'), sTT('selection'))  dsc.putReference(sTT('from'), ref2), dsc.putUnitDouble(sTT('tolerance'), sTT('pixelsUnit'), 1)  executeAction(sTT('make'), dsc)  returned: so every one of them still haven't selected the main subject ... View more

these are the 3 layers I have, which one should I choose before the script? ... View more

tried that with .5 and it took like 10 minutes and returned an error, is there a way to do it without paths? (i think it takes too long) ... View more

Thank for your help JJ, as always. I found it and it's very nice, I looked around the Forums and found these two threads: https://forums.adobe.com/thread/1246514 https://forums.adobe.com/thread/2203096 Based on it I wrote this script: #target photoshop // the script requires to select a folder, then it opens all the TIF or PSD inside it (including subfolders) and saves them as JPG quality 5. var theFolder = Folder.selectDialog("select folder"); var fileandfolderAr = scanSubFolders(theFolder,/\.(tif|psd)$/i);                //var fileandfolderAr = scanSubFolders(topFolder,/\.(jpg|tif|psd|bmp|gif|png|)$/i); var fileList = fileandfolderAr[0]; for(var a = 0 ;a < fileList.length; a++) { var docRef = open(fileList); var Name = app.activeDocument.name                 var savePath = activeDocument.path;         var saveName = File(savePath + "/" + Name)          SaveJPEG(saveName,5);         app.activeDocument.close(SaveOptions.DONOTSAVECHANGES); } function scanSubFolders(tFolder, mask) { // folder object, RegExp or string     var sFolders = new Array();     var allFiles = new Array();     sFolders[0] = tFolder;     for (var j = 0; j < sFolders.length; j++){ // loop through folders                    var procFiles = sFolders .getFiles();         for (var i=0;i<procFiles.length;i++){ // loop through this folder contents             if (procFiles instanceof File ){                 if(mask==undefined) allFiles.push(procFiles);// if no search mask collect all files                 if (procFiles.fullName.search(mask) != -1) allFiles.push(procFiles); // otherwise only those that match mask         }else if (procFiles instanceof Folder){             sFolders.push(procFiles);// store the subfolder             scanSubFolders(procFiles, mask);// search the subfolder          }       }    }    return [allFiles,sFolders]; };     function SaveJPEG(saveFile, jpegQuality){         jpgSaveOptions = new JPEGSaveOptions();         jpgSaveOptions.embedColorProfile = true;         jpgSaveOptions.formatOptions = FormatOptions.STANDARDBASELINE;         jpgSaveOptions.matte = MatteType.NONE;         jpgSaveOptions.quality = jpegQuality;          activeDocument.saveAs(saveFile, jpgSaveOptions, true,Extension.LOWERCASE);          };     ... View more

I use pixels as units anyway, but i'll add this to the beginning of the script anyway ... View more

Works! (Photoshop CC 2019) Thanks!! ... View more

r-bin​, c.pfaffenbichler​ thank you for your replies. I'm on the way to solve it, I made a new path, and recorded the last anchor point in an action, I played the action afterwards and it said "Pick-up Path" I think I have 2 ways to go from here (i need help to execute each one of them since I don't know how) 1. find a script to "Pick-up Path" 2. I've recorded the last anchor point in an action, if I could record the last anchor point and somehow save it for later, for example: write the X,Y position into the Metadata and later delete it, it should work, only question how I record the anchor point or the last anchor point ... View more

Yeah you're right. I expect more of a Windows workaround or plugin like PSD Codec. That's what i'm basically asking for. ... View more

it may be true, but in the same way you can see thumbnail for the psd file in the pictures i attached. Normally one can't see psd preview. I use a software that is called PSD Codec A plugin or workaround that we don't know about might be in the market. There are a lot of upgrades out there, I just saw yesterday a plugin that adds a search bar in Photoshop Actions. ... View more

Filter is great but it's different from sort or group. I'll give you some examples Here is a folder called 'NEF' in Windows 10. it has 2 folders inside it Folder1 and Folder2, I searched for all the NEF files inside the NEF folder so it found NEF files from both folders. Then I grouped by Folder - meaning i will still see files from both folders with a line between the folders: In Adobe Bridge the filter allows to show only one folder at a time: I can also see all the files of NEF folder: but this will not have a line between folders. I can only see one folder at a time or all the files in one place. no "Group by" option. there is also no "sort by focal length option": As I said on the other thread, this is all to distinguish between the angles, I would probably need to look more into Machine Learning option ... View more

Hi gary, Actions, in Photoshop for example, are set of commands that are executed by clicking one button, and therefore make the process a lot easier. Conditionals are IF statements, example: "IF the image is a square THEN apply preset "Square"", i don't think this is possible My end goal is to distinguish between the angles as i told Chuck. I think eventually this will be done with AI & Machine Learning. I have another nice idea that is very easy to do, give Presets keyboard shortcuts! this will be great and will help a lot. Either way, I think ACR really need to take the next step and add Actions. It will give ACR another dimension and take it to the next level. ... View more

Thank you Chuck, i know about this one unfortunately i need to distinguish 6 different angles of my products that were shot with the same camera (same serial number) ... View more

Adobe Bridge is definitely an option, does it have browsing options like Windows: "Group by focal length" and "Group by parent folder" and "Sort by focal length" ? ... View more

wondering if you found a solution to add metadata via Camera Raw? (or plugin as you mentioned) ... View more

It worked! that was awesome! my process is becoming more and more automatic each day thank you my friend ... View more

created a WorkPath, i'm getting this: I wonder if it's possible to make it work without having to create a work path because the script straightens any quadrangle to rectangle did not require a work path, and the script Auto Symmetry Skew did not require a work path. a combination of the two can be pretty awesome ... View more

any chance to bother you to combine the two? this is the auto symmetry one: app.activeDocument.suspendHistory("Skew","skew_straighten_layer()");        function skew_straighten_layer()        {        try {             app.preferences.rulerUnits = Units.PIXELS;              var doc = app.activeDocument;            var layer = app.activeDocument.activeLayer;              selection_from_layer();              var dh = Number(layer.bounds[3].value - layer.bounds[1].value)*0.01;              select_single_line(Number(layer.bounds[1].value) + dh, true, stringIDToTypeID("intersectWith"));              var x0 = Number(doc.selection.bounds[2].value + doc.selection.bounds[0].value)/2;          var y0 = Number(doc.selection.bounds[3].value);            selection_from_layer();              select_single_line(Number(layer.bounds[3].value) - dh, true, stringIDToTypeID("intersectWith"));              var x1 = Number(doc.selection.bounds[2].value + doc.selection.bounds[0].value)/2;          var y1 = Number(doc.selection.bounds[3].value);            doc.selection.deselect();                var w = x1 - x0;            var h = y1 - y0;                var angle;            if (h != 0)               {              angle = Math.atan(w/h) * 180.0 / Math.PI;                  skew(layer, -angle,0, x1, y1);              }          }        catch (e) { alert(e) }        }        function select_single_line(x, h, mode)        {        try            {            if (mode == undefined) mode = stringIDToTypeID("set");                var r = new ActionReference();            r.putProperty(stringIDToTypeID("channel"), stringIDToTypeID("selection"));                var d = new ActionDescriptor();            d.putReference(stringIDToTypeID("target"), r);                var d1 = new ActionDescriptor();            d1.putUnitDouble(stringIDToTypeID(h?"top":"left"), stringIDToTypeID("pixelsUnit"), x);                d.putObject(stringIDToTypeID("to"), stringIDToTypeID(h?"singleRow":"singleColumn"), d1);                executeAction(mode, d, DialogModes.NO);            }        catch (e) { throw(e); }        }        function selection_from_layer()        {        try            {            var r = new ActionReference();            r.putProperty(stringIDToTypeID("channel"), stringIDToTypeID("selection"));                var d = new ActionDescriptor();            d.putReference(stringIDToTypeID("target"), r);                var r = new ActionReference();            r.putEnumerated(stringIDToTypeID("channel"), stringIDToTypeID("channel"), stringIDToTypeID("transparencyEnum"));            d.putReference(stringIDToTypeID("to"), r);                executeAction(stringIDToTypeID("set"), d, DialogModes.NO);            }        catch (e) { throw(e); }        }      function skew(layer, h,v, x, y)      {      try          {          app.activeDocument.activeLayer = layer;            var d1 = new ActionDescriptor();          var d2 = new ActionDescriptor();          var d3 = new ActionDescriptor();          var d4 = new ActionDescriptor();          var r1 = new ActionReference();            r1.putEnumerated( charIDToTypeID( "Lyr " ), charIDToTypeID( "Ordn" ), charIDToTypeID( "Trgt" ) );            d1.putReference( charIDToTypeID( "null" ), r1 );            d1.putEnumerated( charIDToTypeID( "FTcs" ), charIDToTypeID( "QCSt" ), charIDToTypeID( "Qcsi" ) );            if (x != undefined &&  y != undefined)              {                 d2.putUnitDouble( charIDToTypeID( "Hrzn" ), charIDToTypeID( "#Pxl" ), x );              d2.putUnitDouble( charIDToTypeID( "Vrtc" ), charIDToTypeID( "#Pxl" ), y );              d1.putObject( charIDToTypeID( "Pstn" ), charIDToTypeID( "Pnt " ), d2 );              }            d4.putUnitDouble( charIDToTypeID( "Hrzn" ), charIDToTypeID( "#Ang" ), h );          d4.putUnitDouble( charIDToTypeID( "Vrtc" ), charIDToTypeID( "#Ang" ), v );            d1.putObject( charIDToTypeID( "Skew" ), charIDToTypeID( "Pnt " ), d4 );            executeAction( charIDToTypeID( "Trnf" ), d1, DialogModes.NO );          }      catch (e) { throw(e); }        }  and combine it with the jsxbin of this thread. thanks anyway! ... View more

of course i remember the symmetry skew, that's the first thing i thought about, i still have problems combining between the two ... View more

r-bin​ any ideas? ... View more

can we change it from "Layer 1" and "Layer 2" to change active layer to the layer under it? ... View more

I used r-bin​'s Straightens any quadrangle to rectangle script and i straightened a rectangle to the canvas (works perfect): BEFORE AFTER I want to straighten a trapezoid in a way that it will fit up side of the layer to up side of the canvas, and down side of the layer to down side of the canvas automatically, just like that: BEFORE AFTER this is where i got so far (after using r-bin's script): if I use Free Transform > Perspective I can manually change it to: BUT there is a difference in the location of X1 and Y1, X2 and Y2. i made some calculations and i think my logic is right and it's just a matter of execution by script. So let's say this is how my original trapezoid looked like: X1 = 650 X2 = 2650 X2-X1 = 2000 Canvas Width = 4000 and here is how it should be: X1 = 1000 X2 = 3000 X2-X1 = 2000 Canvas Width = 4000 This is how to get there: Y1 should be = (4000-2650+650) /2 = 1000 in other words (Canvas Width - X2 + X1) * 0.5 Y2 should be = 4000-1000 = 3000 in other words Canvas Width -Y1 After I know where my Y1 and Y2 needs to be, there must be an automatic way to: 1. Distort top left corner to Y1 2. Distort top right corner to Y2. Any kind of help will be appreciated ... View more