Small PIL task

Author: cooper-mj

.DS_Store

@@ -166,7 +166,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -310,7 +310,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "These next two functions are really neat, since these are image editing techniques which feature in production image editing software (and apps like Instagram and the iPhone Photos app)!\n",
+    "These next two functions - `brightness` and `contrast` - are really neat, since these are image editing techniques which feature in production image editing software (and apps like Instagram and the iPhone Photos app)!\n",
     "\n",
     "`brightness`: the first thing that you'll note about this function is that, in addition to a NumPy array\n",
     "representing an image, it takes in a factor. Scale the brightness of the image by multiplying every element\n",
@@ -336,6 +336,30 @@
     "save_image(brightness(img, 1.5), \"cs41img_bright.png\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "`brightness_PIL`: it turns out that for many image processing tasks, like this one, NumPy isn't necessarily the best way to go. You'll recall that your implementation of `brightness` likely took a few minutes to run. Can we use Python's image processing libraries to speed up the computation time and make our code even more pythonic?\n",
+    "\n",
+    "Take a moment and check out the `Image.point()` function included with `PIL`. It takes in a lambda (which takes in one argument), then applies that lambda to each point (R, G, B value) in the image. So now that we've implemented our solution to `brightness` above, how might we implement `brightness_PIL`, which makes use of this functionality from `PIL` in order to achieve the same effect?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "def brightness_PIL(img, factor):\n",
+    "    # Implement your function here\n",
+    "    return img\n",
+    "\n",
+    "img = brightness_PIL(Image.open(\"cs41img.png\"), 1.5)\n",
+    "img.save(\"cs41img_bright_PIL.png\")\n"
+   ]
+  },
   {
    "attachments": {
     "SOLcs41img_contrast.png": {
@@ -522,7 +546,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.4"
+   "version": "3.8.0"
   }
  },
  "nbformat": 4,

notebooks/lab-6/cs41img_bright_PIL.png

@@ -347,6 +347,29 @@
     "save_image(brightness(img, 1.5), \"cs41img_bright.png\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "`brightness_PIL`: it turns out that for many image processing tasks, like this one, NumPy isn't necessarily the best way to go. You'll recall that your implementation of `brightness` likely took a few minutes to run. Can we use Python's image processing libraries to speed up the computation time and make our code even more pythonic?\n",
+    "\n",
+    "Take a moment and check out the `Image.point()` function included with `PIL`. It takes in a lambda (which takes in one argument), then applies that lambda to each point (R, G, B value) in the image. So now that we've implemented our solution to `brightness` above, how might we implement `brightness_PIL`, which makes use of this functionality from `PIL` in order to achieve the same effect?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def brightness_PIL(img, factor):\n",
+    "    img = img.point(lambda p: min(p*factor, 255))\n",
+    "    return img\n",
+    "\n",
+    "img = brightness_PIL(Image.open(\"cs41img.png\"), 1.5)\n",
+    "img.save(\"cs41img_bright_PIL.png\")"
+   ]
+  },
   {
    "attachments": {
     "SOLcs41img_contrast.png": {