• 解决 Faster R-CNN 图片中框不在一张图片上显示的问题


    解决 Faster R-CNN 图片中框不在一张图片上显示的问题

    发现问题

    在使用demo.py的时候,选取测试用的图片,放到demo,然后修改demo.py中对应的图片名称,然后进行测试:

    发现:图片中被框出来的部分并没有完全到一张图片上去,经过多张图片的测试,可以发现,并不是一张图片上一个框,而是按照类别进行的划分,即:每一类一张图片

    如何解决这个问题?

    原先的画图部分主要在这里:

    def vis_detections(im, class_name, dets, thresh=0.5):
        """Draw detected bounding boxes."""
        inds = np.where(dets[:, -1] >= thresh)[0]
        if len(inds) == 0:
            return
    
        im = im[:, :, (2, 1, 0)]
        fig, ax = plt.subplots(figsize=(12, 12))
        ax.imshow(im, aspect='equal')
        for i in inds:
            bbox = dets[i, :4]
            score = dets[i, -1]
    
            ax.add_patch(
                plt.Rectangle((bbox[0], bbox[1]),
                              bbox[2] - bbox[0],
                              bbox[3] - bbox[1], fill=False,
                              edgecolor='red', linewidth=3.5)
                )
            ax.text(bbox[0], bbox[1] - 2,
                    '{:s} {:.3f}'.format(class_name, score),
                    bbox=dict(facecolor='blue', alpha=0.5),
                    fontsize=14, color='white')
    
        ax.set_title(('{} detections with '
                      'p({} | box) >= {:.1f}').format(class_name, class_name,
                                                      thresh),
                      fontsize=14)
        plt.axis('off')
        plt.tight_layout()
        plt.draw()
    
    def demo(net, image_name):
        """Detect object classes in an image using pre-computed object proposals."""
    
        # Load the demo image
        im_file = os.path.join(cfg.DATA_DIR, 'demo', image_name)
        im = cv2.imread(im_file)
    
        # Detect all object classes and regress object bounds
        timer = Timer()
        timer.tic()
        scores, boxes = im_detect(net, im)
        timer.toc()
        print ('Detection took {:.3f}s for '
               '{:d} object proposals').format(timer.total_time, boxes.shape[0])
    
        # Visualize detections for each class
        CONF_THRESH = 0.8
        NMS_THRESH = 0.3
        for cls_ind, cls in enumerate(CLASSES[1:]):
            cls_ind += 1 # because we skipped background
            cls_boxes = boxes[:, 4*cls_ind:4*(cls_ind + 1)]
            cls_scores = scores[:, cls_ind]
            dets = np.hstack((cls_boxes,
                              cls_scores[:, np.newaxis])).astype(np.float32)
            keep = nms(dets, NMS_THRESH)
            dets = dets[keep, :]
            vis_detections(im, cls, dets, thresh=CONF_THRESH)
    

    修改为:

    # 将检测可视化
    def vis_detections(ax,im, class_name, dets, thresh=0.5):
        """Draw detected bounding boxes."""
        #print("+_+")
        #print(class_name,dets,thresh)
        inds = np.where(dets[:, -1] >= thresh)[0]
        print("!!!")
        #print(inds) # 是否检测出来东西,如果有的话为0如果没有为空
        if len(inds) == 0:
            return
        #print(im.shape)  # 4000 6000 3
        #调整通道顺序,如果不调整通道顺序,图像就不正常
    
        for i in inds:
            bbox = dets[i, :4]
            score = dets[i, -1]
            #print(bbox[0],bbox[1],bbox[2],bbox[3])
            print("add one patch")
            ax.add_patch(
                plt.Rectangle((bbox[0], bbox[1]),
                              bbox[2] - bbox[0],
                              bbox[3] - bbox[1], fill=False,
                              edgecolor='red', linewidth=2)
                )
            ax.text(bbox[0], bbox[1] - 2,
                    '{:s} {:.3f}'.format(class_name, score),
                    bbox=dict(facecolor='white', alpha=0.9),
                    fontsize=8, color='black')
        ax.set_title(('{} detections with '
                      'p({} | box) >= {:.1f}').format(class_name, class_name,thresh),fontsize=12)
    
    
    def demo(sess, net, image_name):
        """Detect object classes in an image using pre-computed object proposals."""
    
        # Load the demo image
        im_file = os.path.join(cfg.FLAGS2["data_dir"], 'demo', image_name)
        im = cv2.imread(im_file)
    
        # Detect all object classes and regress object bounds
        timer = Timer()
    
        timer.tic()
        # detect the picture to find score and boxes
        scores, boxes = im_detect(sess, net, im)
        # 检测主体部分,在这里加上save_feature_picture
        # 这里的net内容是vgg
    
        timer.toc()
    
        print('Detection took {:.3f}s for {:d} object proposals'.format(timer.total_time, boxes.shape[0]))
    
        # Visualize detections for each class
        CONF_THRESH = 0.8
        NMS_THRESH = 0.3
    
        im = im[:, :, (2, 1, 0)]
        fig, ax = plt.subplots(figsize=(10,10))
        ax.imshow(im, aspect='equal')
    
        for cls_ind, cls in enumerate(CLASSES[1:]):
            cls_ind += 1  # because we skipped background
            cls_boxes = boxes[:, 4 * cls_ind:4 * (cls_ind + 1)]
            cls_scores = scores[:, cls_ind]
            dets = np.hstack((cls_boxes,
                              cls_scores[:, np.newaxis])).astype(np.float32)
            keep = nms(dets, NMS_THRESH)
            dets = dets[keep, :]
            vis_detections(ax,im, cls, dets, thresh=CONF_THRESH)
            plt.draw()
    

    点拨:一开始的时候我也没有发现这两个有什么区别,后来发现图片是按照类别进行区分的以后,就可以看出,demo函数中是按照类别进行划分的,所以只要修改plt位置,将plt从vis_detections中放到demo中,这样所有类都会花在同一个plt中,而不会分开了,这样就解决了这个问题。

    参考issues

    How detect multiple object on same picture?

  • 相关阅读:
    CUDA C Best Practices Guide 在线教程学习笔记 Part 1
    0_Simple__simpleCallback
    0_Simple__simpleAtomicIntrinsics + 0_Simple__simpleAtomicIntrinsics_nvrtc
    0_Simple__simpleAssert + 0_Simple__simpleAssert_nvrtc
    0_Simple__matrixMulDrv
    0_Simple__matrixMulCUBLAS
    0_Simple__matrixMul + 0_Simple__matrixMul_nvrtc
    0_Simple__inlinePTX + 0_Simple__inlinePTX_nvrtc
    0_Simple__fp16ScalarProduct
    0_Simple__cudaOpenMP
  • 原文地址:https://www.cnblogs.com/pprp/p/9530130.html
Copyright © 2020-2023  润新知